diff --git a/CHANGELOG.md b/CHANGELOG.md
index 84b262257..090246ce9 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -36,7 +36,7 @@ All notable changes to OriginWeave are documented in this file. The format follo
 - Recorded the verified pointer-reply connection repair and recovery tests, keeping pending hosted checks, remaining authority gaps and browser acceptance separate.
 - Recorded the published text-entry connection safeguards and their test evidence, separating queued hosted checks and remaining pointer/status repairs from released browser behavior.
 - Separate the current maintenance queue and source lineage from preserved historical observations, and record the text-entry session and pending-request safeguards without claiming released browser behavior.
-- Current delivery inventory: 131 open pull requests (14 ready, 117 draft); 14 open non-PR issues. Observed 2026-09-09; source acceptance remains revision-specific.
+- Current delivery inventory: 130 open pull requests (12 ready, 118 draft); 14 open non-PR issues. Observed 2026-09-09 12:57 UTC; source acceptance remains revision-specific.
 - Clarified that the active presentation-evidence harness is implemented while its failed browser run does not establish product runtime or release acceptance.
 - Corrected historical checkpoint labels and made the current inventory check reject stale counts; recorded verified text-input parent adoption without claiming browser execution.
 - Recorded the executed session-isolation repair and its verified adoption, separating complete local checks and PR visual inspection from pending hosted checks and unreleased browser acceptance.
diff --git a/docs/evidence/CHANGELOG-through-2026-09-09.md b/docs/evidence/CHANGELOG-through-2026-09-09.md
new file mode 100644
index 000000000..84b262257
--- /dev/null
+++ b/docs/evidence/CHANGELOG-through-2026-09-09.md
@@ -0,0 +1,202 @@
+# Changelog
+
+All notable changes to OriginWeave are documented in this file. The format follows Keep a Changelog, and releases use Semantic Versioning.
+
+## [Unreleased]
+
+- Recorded #298/#305's repaired reduced-motion command contract while keeping
+  its Draft-only checks and missing Chromium runtime evidence explicit.
+
+- Repaired the dated product-gap checkpoint contract so historical verification
+  remains bounded after a newer live cut is added.
+
+- Replaced the truncated inline merge-evidence command copy with its canonical executable collector and kept the collector's exact-head contract under test.
+
+- Recorded #255 exact-head hosted success and #293's narrow standard-BiDi capability, typed command planning, dated-TR provenance pin, and explicit viewport/DPR cleanup intent, preserving exact-head visual, rustdoc-view, transport, and real-browser evidence gaps.
+
+- Recorded the server Close-role RED-to-GREEN repair with exact local coverage and visual evidence, keeping queued hosted checks and release acceptance separate.
+
+- Recorded completed local reply-verification checks, visual inspection and exact-head hosted success, distinguishing clean Draft state from protected-main and release acceptance.
+
+- Refreshed transport-closure evidence with completed predecessor CI, the separately pending test-integrity repair, and the remaining deadline and control-frame behavior gaps.
+
+- Recorded published session-end reply protections, their visual and local verification, and the new browser-presentation evidence gap without treating queued checks or retained replies as release or live-stream acceptance.
+- Bound each new delivery-evidence row to its own publication state, revision, parent and coverage so one row cannot hide another row's missing evidence.
+
+- Record the published semantic-action adoption stack and its local verification, visual evidence and unmet release prerequisites without treating open pull requests as shipped behavior.
+
+- Record verified maintenance lessons for resumable checks, rendered documentation links, parent-preserving repairs and release-readiness checks without exposing credentials.
+
+- Recorded published observation safeguards and original-connection recovery with separate local, queued hosted and visual evidence; full browser acceptance remains unfinished.
+- Recorded published text-response and fixed-observation progress with separate local, hosted and visual evidence; browser outcome verification remains unfinished.
+- Recorded published click and text-entry safeguards with their own test, coverage and visual evidence, keeping queued hosted checks and remaining browser outcomes separate from delivery.
+- Recorded the published subscription descendants and completed source visual inspection, correcting the stale unresolved-review count while keeping queued checks, required approval and unreleased browser acceptance distinct. Compact linked revisions keep the new evidence table readable without hiding coverage off-screen.
+- Recorded the verified subscription-reply connection repair, preserving queued hosted checks, the locked-screen visual-inspection gap and remaining authority and delivery work.
+- Recorded verified navigation and subscription adoption of the click-reply repair, distinguishing local checks and rendered evidence from queued hosted acceptance and remaining provenance gaps.
+- Recorded the verified pointer-reply connection repair and recovery tests, keeping pending hosted checks, remaining authority gaps and browser acceptance separate.
+- Recorded the published text-entry connection safeguards and their test evidence, separating queued hosted checks and remaining pointer/status repairs from released browser behavior.
+- Separate the current maintenance queue and source lineage from preserved historical observations, and record the text-entry session and pending-request safeguards without claiming released browser behavior.
+- Current delivery inventory: 131 open pull requests (14 ready, 117 draft); 14 open non-PR issues. Observed 2026-09-09; source acceptance remains revision-specific.
+- Clarified that the active presentation-evidence harness is implemented while its failed browser run does not establish product runtime or release acceptance.
+- Corrected historical checkpoint labels and made the current inventory check reject stale counts; recorded verified text-input parent adoption without claiming browser execution.
+- Recorded the executed session-isolation repair and its verified adoption, separating complete local checks and PR visual inspection from pending hosted checks and unreleased browser acceptance.
+- Recorded the current click safeguards, restored regression coverage and completed PR visual inspection, keeping the separately owned parent repair and queued hosted checks distinct from delivery.
+- Recorded subscription-shutdown safeguards and their complete local verification, keeping queued hosted checks and the locked-screen visual-inspection gap explicit.
+- Recorded the repeated-response safeguard, its complete local verification and visual review, separating successful predecessor checks from the new revision's pending checks.
+- Recorded the original-browser-state subscription repair and its final verification, keeping pending hosted checks and unfinished browser acceptance separate.
+- Recorded the published subscription integration and its complete local verification, preserving earlier failures and the separate hosted-check and release requirements.
+- Recorded failed quality checks on the current subscription revision without transferring passing results from an unpublished repair.
+- Corrected the browser acceptance status after hosted trials failed, and separated verified unpublished subscription safeguards from another task's ongoing changes.
+- Refreshed the live delivery baseline with sandbox-parent adoption, subscription request-identity repair, current review gates and explicit limits on browser acceptance evidence.
+- Separated the September 6 verified queue from prior measurements, recording restored-foundation workflow failures, current successor CI results, completed sandbox doctoring, and unreleased runtime/central-owner prerequisites.
+- Recorded verified #261/#277 parent adoption, the reproduced subscription-deadline slot repair, current #288 sandbox-source verification, and the central report-classification source repair; retained hosted, review, runtime, and independent-verification limits and the REST refresh boundary.
+- Recorded #50's reproduced formatting repair and verified current-parent integrations through #255–#260, preserving their exact local test/coverage evidence, child-source ownership, warning and runtime limits, and remaining hosted/approval gates.
+- Corrected #50's live head after the exit sweep found its newer default-port regression, keeping earlier passing results attached to their tested predecessors.
+- Recorded #254 parent adoption, #148's crash-launch sandbox repair and #150's shared-observer test reconciliation, including complete local verification and matched-source Linux runs; separated the reproduced central scan-classifier defect from a gateway failure and hosted execution-limit cancellation.
+- Recorded #253's verified current-parent integration while retaining its unresolved teardown-evidence finding, downstream repair ownership and instrumentation warning; local passing checks do not establish operational completion.
+- Recorded #252's verified current-parent integration and #147's test-only repair from two reproduced failures to all 228 passing tests, preserving the earlier RED history, ordinary-pass deadline finding and instrumentation warning.
+- Recorded #251's verified parent integration and #147's restored parent evidence with two remaining test-reconciliation failures; distinguished the externally closed temporary synthesis from product acceptance and refreshed the live inventory.
+- Recorded #255's new owner-verified repair separately from its prior quality failures, retained the real branch-instrumentation warning, and identified #147's lossy parent adoption with a failing replay of six omitted regressions.
+- Recorded #144/#145 parent adoption, #146's reproduced cleanup-evidence fixes and preserved concurrent integration, with complete local verification separated from queued hosted compatibility; refreshed #287's executed Strix scan and failed Noema gateway retry evidence without claiming recovery or approval.
+- Repaired #248's uncollected release-record test and obsolete merged-parent dependency; recorded 142 passing Python contracts, full local Rust gates and exact coverage while preserving Draft and hosted-check boundaries.
+- Recorded #246's verified parent integration and exact local coverage, plus #141's unchanged-head cleanup review and two resolved informational findings; hosted and protected-main acceptance remain separate.
+- Recorded the shared opening-exchange fixture repair in #242 and its non-force adoption by #243, with full local gates, exact coverage and queued hosted checks kept distinct.
+- Recorded #255's passing received-connection regressions alongside its remaining exact-snapshot quality failures, and #139's verified documentation recovery with three resolved review findings.
+- Corrected #285's stale native-CI pending claim to exact-head Rust and coverage success, and recorded supported replay admission for its three cancelled security workflows without claiming scan results or lifecycle-trigger repair.
+- Recorded #255's remaining cross-connection response substitution after the sender/closure repair, with a failing real-socket regression and the prerequisite incoming-message binding.
+- Separated #219/#240 terminal Noema gateway failures from queued central CodeQL scans; one failed-job retry per unchanged head is admitted, but no provider recovery or passing review is claimed.
+- Recorded #50's current documentation-repair head and resolved review, preserving predecessor-only full-coverage evidence and the new head's independent hosted-check requirement.
+- Recorded a reproduced cross-connection teardown-evidence gap in #254/#255 and its prerequisite repair order; another connection's closure must not be reported as closure of the acknowledged connection.
+- Made the changelog queue-consistency check compare the current baseline counts directly, preventing a duplicated older count from rejecting a verified Ready-to-Draft transition.
+- Converted PR #290 back to Draft after exact review showed its #245-relative MV3 workflow would drop protected #286 `converted_to_draft`/`closed` lifecycle coverage and the closed-event fail-closed guard; the valid concurrency/manual-dispatch intent remains assigned to canonical workflow owner #212.
+- Recorded the #253 teardown-evidence gap and corrected #283's stale #282 predecessor-base classification; raw caller claims and stale parent compares cannot become current completion evidence.
+- Refreshed the active WebDriver BiDi and semantic-action stack evidence through exact heads #195, #242, #93, #95, #96, #101, #102, and #103, including macOS fixture-race repairs and fail-closed policy, dispatch, disabled-state, and current-observation boundaries.
+- Recorded PR #285's Ready-event materialization gap: repository-native exact-head CI restarted, while the earlier central Security, Semgrep, and CodeQL runs remained cancelled and no fresh required-workflow replacements appeared.
+- Refreshed the live product-gap baseline with PR #37's RFC 9110 `205 Reset Content` repair head and exact local verification evidence.
+- Recorded PR #284's rule-suite `3948421709` administrative bypass into protected `main` at `4ed08bfa7c063fc7f2ef9278ee8d281887b8296b`; post-merge checks remain independently required and cannot retroactively turn the bypass into policy-compliant integration.
+- Refreshed the product-gap queue to 125 open pull requests (12 ready, 113 draft) and 13 open non-PR issues on 2026-09-05; queue movement is active-PR evidence, not protected-main shipment.
+- Recorded the current Ready-root heads for #37, #50, #166, #219, #220, #229, #240, #272, #274, #285, and #287; Draft #290 remains a #245-dependent workflow-owner candidate, #238's moving self-reference is delegated to live PR metadata, hosted exact-head checks remain non-terminal, and counted approval remains absent.
+- Corrected the retained PR #219 exact-head regression anchor and separated the shipped `tools/call` policy binding from the core-only `tools/list` discovery contract.
+- Refreshed the #279 evidence to the current #282/#283 heads, including complete raw-diff object identities and rename/copy similarity binding to blob identity.
+
+### Added
+
+- Recorded the published browser-status repair with exact local coverage and pending hosted checks; added verified coverage-repair guidance to `AGENTS.md` without changing release or approval gates.
+- Recorded the verified #250 status-response parent adoption, #144's conservative process-set review limits and #287's authenticated CodeQL dispatch handoff, without promoting cancelled or pending hosted evidence to acceptance.
+- Recorded the verified #249 parent integration, #142/#143 failure-cleanup review evidence, and #255's remaining formatting/lint/coverage failures, keeping local results distinct from pending browser/hosted acceptance.
+- Revalidated the product-gap queue at 125 open pull requests (12 ready, 113 draft) and 13 open non-PR issues on 2026-09-05; protected `main` is `87c4daa1830bac5a5228b6036752ad5633232085`, presentation-identity PR #229 is Ready at `024f63690cf05cfe6f0d4a430f0e18ea8fd2c4d6`, WebDriver BiDi document-advance PR #260 is Draft at `3a651967c421f77088fe25e86a63faae295390b3`, denial-error PR #166 is Ready at `e84a1a2cc82b1c666218efd441da97849f47b8c2`, enterprise-approval PR #220 is Ready at `e545b94e1de499b96b867694f80ac04ad247becd`, JSON-envelope PR #247 has merged into its unprotected parent and successor #248 is Draft at `b386f17c4826adabebda084bff2fba35aee94dd0`, DDD/MCP repair #272 is Ready at `b1cae8ad1cbd8eb6992037c830aea30b9aa436b3`, governance repair #285 is Ready at `f455c2cd64b3dd3f027c91d396103792a205ddd0`, subscription repair #277 is Draft at `01038ba71fb276426cc67f90a91a3c431e194db5`, classifier foundation #287 is Ready at `af83c40dd2990a03064a92ca75430a9cc400f098`, workflow-free Agent Task successor #288 is Draft at `39e36256651f62940ec3ca6149067f0cfcb2285a`, and workflow-owner candidate #290 is Draft at `ebeefcd534db4324498fdb18046ebc6255ddcdf2`. Current-head checks remain independently required; no active-PR work is claimed as protected-main shipment.
+- Recorded PR #67's exact-head squash merge into unprotected parent #64 at `5021d142583cb5a8e393248048bb824762a98056`; #64's hosted checks restarted on its new exact head, so neither child nor parent evidence is promoted to protected-main delivery.
+- Recorded PR #217's squash merge into the unprotected #210 feature parent at `66f360ccac5cec60c72222cc79d58e39f6f00088`; #210's current exact-head checks are regenerating, so the merged child evidence was not promoted to protected-main delivery.
+- Recorded PR #210's superseded exact-head failures at `bea65643109449d63d367a35b8d9bf327ee7cb2c`: OpenCode had no current-head verdict and Strix exhausted three provider HTTP 500 attempts without a vulnerability report; its direct attempt-2 rerun (`33172708455` / `98915847518`) reached the same fail-closed result, so no failure was promoted to passing evidence for the new head.
+- Recorded the current-head OpenCode dispatch authorization failure for PR #46 (`33192478312` / `98921183278`) and the immediately preceding PR #238 head (`33192483900` / `98921203971`): the central validator rejected the maintainer actor because it was not the configured scheduler identity; no review verdict or approval was synthesized.
+- Added an enforcing traceability contract for the protected-main MCP `tools/list` maturity claim and indexed its executable evidence dossier.
+- Clarified the 2026-08-28 gap snapshot's prior queue observation and separated open issues from governance signals.
+- Historical 2026-08-28 product-gap snapshot: protected-main `542ca1e9`, 110 open pull requests (26 ready, 84 draft), 11 open issues, current exact base/head pairs for representative PRs including #70, #82, #152, #37, #237, and #239, PR #170's merged `tools/list` contract, the active counted-approval gate, and the unchanged Phase 1 browser gap.
+- Historical 2026-08-28 baseline refresh: current queue and exact-head evidence were explicitly separated from the historical 2026-08-26 snapshot so stale delivery evidence cannot be promoted.
+- Hardened the product-completion documentation contract by classifying current-snapshot count markers explicitly instead of routing them by substring heuristics.
+- Corrected the baseline's #72/#73 exact SHA evidence to full 40-character values and tracked PR #238's current head.
+- Historical predecessor: PR #210 Ready head `d83748a70bd1b16dbfec46007fe02989ba6ce188` was superseded by exact Ready head `0341079331f9cea669eb9a5cc21842fd6027431e`; the earlier entry recorded its nested credential-query defense and current-head hosted-check regeneration.
+- Refreshed the baseline's #210 WARC row to exact Ready head `0341079331f9cea669eb9a5cc21842fd6027431e`, including encoded-control rejection, and corrected required-workflow provenance to the central `.github` repository with exact OpenCode/Strix failure evidence without weakening any gate.
+- Recorded the exact #238 OpenCode model-unavailable rerun (`33182772296`) and central scheduler dispatch (`33182749298`), and linked the central Rust coverage repair PR #1391 without promoting review-tool output to approval.
+- Removed the self-referential #238 exact-head row from the baseline; the live PR metadata and body remain authoritative because every documentation update necessarily creates a new branch head.
+
+- Added `originweave_core::release_acceptance`, a deterministic fail-closed benchmark release-decision contract that requires one authoritative result for every mandatory suite, bounds explicit buyer-visible limitations, rejects duplicate limitation claim identities, and rejects non-canonical surrounding whitespace rather than normalizing it into an alternate claim spelling.
+- Refreshed the product and technical gap baseline with the 2026-08-24 live inventory: 158 open pull requests (44 ready, 114 draft), refreshed exact base/head evidence for the #208–#222 release, enterprise-approval, BAP, and WARC/PROV chains, the governance issue additions #212 and #215, and a required-check provider-failure record for the fail-closed Strix re-dispatches on #208/#218/#220.
+- Added a dated product and technical gap baseline that separates protected-main implementation truth, active pull-request evidence, live review/check blockers, and the next buyer-visible Phase 1 acceptance work.
+- Refreshed the product and technical gap baseline with the current open-PR inventory and exact-head evidence for the newest Chromium, BAP, extraction, WARC, and idempotency slices.
+- Bound explicit extension-to-Agent grants to exclusive trusted-time expiry in addition to extension identity, session, browsing context, and canonical origin, so a same-origin grant cannot be reused at or after the deadline.
+- Bound explicit extension-to-Agent grants to the exact canonical origin in addition to extension identity, session, and browsing context, so a same-session navigation or port change cannot reuse the grant.
+- Rust workspace for independently reusable core, policy, destination, network, TLS, resource, and evidence modules.
+- Canonical HTTPS and loopback-origin boundary with case-normalized schemes and hosts, default-port normalization, IPv4/IPv6 handling, browser-special numeric-host rejection, and explicit malformed-input errors.
+- Typed browser actions, capabilities, risk classes, execution modes, robots decisions, secret-delivery contracts, immutable canonical action-intent digests, and intent-bound approval scopes.
+- Protected main now contains deterministic MCP `2026-07-28` stateless `tools/call` routing with bounded method/tool names, a single reviewed tool-to-action registry shared by routing and discovery metadata, and fail-closed policy binding that grants no ambient authority. The complete MCP adapter, transport serialization, discovery response handling, OAuth, browser I/O, and persistence remain planned.
+- Protected main now includes the conservative MCP `2026-07-28` `tools/list` discovery metadata merged through PR #170: `resultType = complete`, zero freshness, private cache scope, no continuation cursor, per-request protocol/client-capability admission, and bounded protocol-version and method metadata validated before cross-field comparison. This grants no browser, network, secret, approval, or Agent authority and does not complete the MCP adapter.
+- Deterministic fail-closed policy evaluation for untrusted instructions, origin grants, crawler restrictions, execution-mode and purpose consistency, approvals, and brokered secrets.
+- Fail-closed resolved-destination policy with IPv4/IPv6 special-purpose and reviewed cloud-platform endpoint classification, IPv4-mapped canonicalization, explicit class grants, non-empty origin-bound DNS snapshots capped at 256 resolver addresses, concrete connection pinning, DNS-set expansion detection, and per-hop redirect reauthorization.
+- Bounded resolution-freshness authority with trusted monotonic approval time, capped non-zero validity, half-open use windows, non-expanding revalidation, and credential-free authorization timestamps.
+- Direct-only `originweave-network` TCP boundary with explicit canonical `SocketAddr` authority, zero IPv6 flow and scope metadata unless separately modeled, a non-cloneable single-use plan, a 30-second per-attempt timeout ceiling, at most four attempts, exact `peer_addr` verification before stream exposure, and no hostname re-resolution or ambient proxy inheritance.
+- Authenticated `originweave-tls` service-identity boundary that consumes an existing verified TCP stream, requires exact TLS-origin and transport-origin equality, derives RFC 9525 DNS or literal-IP reference identity only from the canonical HTTPS origin, validates WebPKI with explicit roots and fixed time, permits only TLS 1.2 and TLS 1.3, and never reconnects or resolves.
+- Bounded TLS policy for total handshake time, ALPN identifiers, trust-root count and bytes, and server-presented certificate count and bytes, with explicit optional-versus-required ALPN behavior and `NotConfigured` revocation evidence.
+- Deterministic TLS revocation-material freshness authority with a strict signed `thisUpdate`→`nextUpdate` half-open window and typed invalid-window, not-yet-valid, and stale failures, without claiming OCSP/CRL acquisition, cryptographic validation, or certificate revocation status.
+- Credential-free TLS evidence containing canonical origin, TCP peers, reference identity, TLS version, cipher-suite identifier, selected ALPN or explicit absence, leaf certificate and SPKI hashes, server-presented certificate hashes and bounds, trust-bundle identity and hash, validity interval, fixed verification time, revocation configuration, and measured handshake duration.
+- Credential-free sensitive-handle lifecycle evidence binds issuance, exclusive expiry, bounded uses, observed resolution count, and revocation to the exact credential-free `OpaqueHandleOnly` sensitive-access receipt, preserving tenant, actor, task, field set, purpose, destination, classification, policy version, and decision time without storing opaque handle tokens or protected values.
+- Credential-free connection and redirect evidence containing canonical addresses, destination classes, target digests, hop numbers, and approved-address counts.
+- Credential-free verified TCP evidence containing the logical origin, requested socket, observed peer, destination class, successful attempt number, and per-attempt timeout.
+- Standard `Display` and `std::error::Error` contracts for destination, redirect, digest, direct-network, TLS, and resource-budget failures, including preserved destination-policy, rustls, and operating-system sources where applicable.
+- Real loopback TCP integration proof plus deterministic timeout, refusal, retry, peer-inspection, peer-mismatch, canonicalization, IPv6 metadata, and single-use replay tests.
+- Real loopback rustls integration covering trusted DNS SAN, Common-Name fallback rejection, wrong-name and untrusted-root rejection, fixed-time expiry and not-yet-valid failures, exact IPv4 and IPv6 SANs, TLS 1.2/TLS 1.3, required and optional ALPN, and transport-origin binding.
+- Cumulative interactive-first RAM, VRAM, batch, local-model, admission, pause, and compositor-pressure mitigation plans, including active-consumer reduction at exact hard limits.
+- Universally value-redacted network evidence with explicit path, metadata, and provenance bounds; ambiguous path rejection; validated source URLs; lowercase SHA-256 identifiers; and verification state.
+- Versioned schema-bound extraction contracts with bounded identifiers and field counts, typed value/cardinality metadata, explicit duplicate-free reviewed source channels, fail-closed schema validation, and deterministic `Display`/`std::error::Error` contracts for public schema failures.
+- Rust 1.97.1 build contract, strict Clippy and rustdoc gates, and exact production function, line, region, and branch coverage enforcement.
+- Hourly bounded OpenCode product-development workflow using `NVIDIA_NIM_API_KEY`, an unprivileged disposable workspace, loopback-only model broker, independently verified patches, and publication through a dedicated `OPENCODE_PR_TOKEN` that cannot review or merge.
+- Architecture, agent, security, contribution, research, database naming, roadmap, quality-gate, and TLS service-identity ADR documentation.
+- Resumable BAP lifecycle restoration with monotonic sequence recovery and fail-closed sequence exhaustion.
+- Authoritative product documentation graph spanning PRD, TRD, ADR lifecycle/index, product-wide UML, conceptual ERD, requirement/decision traceability, threat modeling, product-wide test strategy, operability, API/protocol, release/rollback, and current primary-source standards doctoring, with machine-checkable repository contracts that keep conversation-derived future work distinct from protected-main implementation claims.
+- Purpose-bound data-governance and privacy baseline that rejects both blanket masking and ambient raw-value propagation, defines field-scoped just-in-time disclosure, opaque-handle/trusted-broker boundaries, model/provider/region policy, retention/deletion/residency/break-glass controls, truthful CSAP/SOC 2 readiness language, and machine-checkable documentation contracts without inventing an OriginWeave-owned production database.
+- Proposed product-wide target-architecture ADRs for the Rust control plane, isolated execution modes, typed actions, semantic observation/stale-node authority, prompt-injection and secret separation, resource-governor priority, provenance evidence, browser/protocol adapters, crawler policy, and hourly automation operational closure; these remain Proposed rather than shipped claims until protected review and merge.
+
+### Changed
+
+- Updated the delivery checkpoint for bounded repeated keepalive handling, retaining the distinction between local verification, hosted acceptance and operational completion.
+
+- Refreshed the delivery checkpoint for completed local shutdown-deadline and reserved-code repairs while keeping hosted acceptance and release gaps explicit.
+
+- Made the open non-PR issue count reproducible in the baseline evidence procedure and corrected the historical exact heads for PRs #53 and #217.
+- Revalidated the active ruleset inventory at 7 required workflows from the live branch-rules API by adding `codeql-pr`; `close-empty-pr` remains useful repository automation but is not a current protected-main requirement.
+
+- Aligned the hourly product-development branch-coverage toolchain and its one-shot materializer with the reviewed `nightly-2026-08-18` pin, and corrected the official Dependabot Rust-toolchain reference.
+- Separated logical origin authority from resolved network destination authority; an origin grant no longer implies permission to connect to every resolver result.
+- Separated resolved-address authorization from direct transport evidence; an approved IP now becomes a usable stream only after the operating system reports the exact requested IP and port.
+- Separated exact TCP peer proof from authenticated TLS service identity; an observed peer becomes an authenticated HTTPS stream only after explicit-root, fixed-time, SAN-bound WebPKI verification over that same stream.
+- Replaced single resource-pressure directives with a cumulative mitigation plan so simultaneous RAM, VRAM, frame, model, and admission pressure cannot discard required actions.
+- Changed generic network capture from finite deny-lists or safe-name allow-lists to unconditional value redaction. Typed metadata values and bodies now require a separate schema-specific capture contract.
+- Updated the first Chromium slice to distinguish implemented origin, destination, direct TCP, and TLS identity kernels from the remaining trusted DNS adapter, proxy/PAC, HTTP budget, MIME, download, and Chromium integration required before safe navigation can be claimed.
+- Separated hourly product PR publication authority from the organization review and merge system, and added live default-branch and release-blocker rechecks immediately before publication.
+- Made the agent-development contract work-conserving: completing one bounded slice, RCA, review request, check, merge, or documentation change is an intermediate state; maintenance must return to the live queue, treat waits as item-local, and perform a mandatory exit sweep before terminating while executable OriginWeave work remains.
+- Hardened the dated baseline evidence collector with fail-fast isolated artifacts, paginated branch and collaborator rules, and post-collection exact-head revalidation.
+- Flattened every paginated workflow-run page in the baseline merge verdict so exact-head evidence cannot silently discard later runs.
+- Hardened the baseline evidence procedure with exact-head legacy status and workflow-run capture, counted approval binding, required-workflow recording, merge verdict artifacts, and bounded moving-head retries.
+- Moved autonomous-agent Cargo targets and Python bytecode caches outside the proposed source tree and prefetched locked Cargo dependencies for offline verification.
+- Updated research doctoring to pin Chromium canonicalizer evidence to an immutable revision, add RFC 9293, RFC 5280, RFC 8446, RFC 9525, rustls 0.23.42, and Rust `TcpStream` evidence, distinguish the April 2026 Fugu beta from the June 2026 release, and treat vendor benchmark claims as first-party evidence rather than independent validation.
+- Tightened the product-baseline contract so the BiDi opening path and VPN/profile evidence retain their explicit not-shipped status within their own documentation sections.
+- Refreshed the product and technical gap baseline against the 2026-08-21 live inventory: 150 open pull requests, 110 drafts, and the new hardened-runner/MV3 evidence gap issue #206.
+- Tightened the baseline completion-gap contract so superseded inventory counts (including the 2026-08-21 150/40/110 snapshot) can no longer pass as current evidence.
+- Refreshed the baseline's merge-authority statement to the live ruleset: two approving reviews are required, while the collaborator inventory still contains only the solo maintainer.
+- Corrected the baseline evidence collector to flatten every paginated input, apply current reviewer and last-push approval semantics, and discard verdicts when either the PR head or base moves.
+
+### Security
+
+- Explicit proxy server identifiers require ASCII decimal port tokens before numeric range parsing, preventing Rust-specific leading-plus spellings from widening proxy authority.
+- Raw page content cannot become a trusted instruction.
+- Raw secrets are rejected and secret-capable actions require an opaque broker handle.
+- Crawler mode is read-only, must pair with the public-crawl purpose, and fails closed without an applicable robots-policy decision.
+- State-changing actions are same-origin by default.
+- R3 and R4 approvals are bound to the exact action, target origin, and immutable digest of the complete canonical action intent; R5 legal consent is non-delegable.
+- Shortened, integer, hexadecimal, and legacy octal-looking IPv4 host spellings are rejected so the policy origin cannot diverge from Chromium host interpretation.
+- IPv4-mapped IPv6 is canonicalized before destination classification and pin comparison so mapped private or loopback addresses cannot bypass IPv4 policy.
+- The default destination policy permits only public addresses and denies unspecified, loopback, private, shared, link-local, metadata, documentation, benchmarking, multicast, broadcast, transition, and protocol-reserved destinations.
+- Azure platform IP `168.63.129.16` and Amazon EKS Pod Identity endpoints `169.254.170.23` and `fd00:ec2::23` are classified as metadata or platform services before broader public, link-local, or unique-local rules.
+- Resolver answers are rejected when empty or larger than 256 addresses, preventing an unbounded resolver response from entering policy state.
+- `localhost` may approve only loopback addresses, while literal IPv4 and IPv6 origins may approve only the exact canonical address encoded in the origin.
+- Resolver answers must remain a non-empty subset of the origin-bound approved address set; any newly introduced address fails closed as a possible DNS-rebinding event.
+- Every redirect rechecks target-origin authority, target-bound resolution, HTTPS downgrade, complete-target cycle state, and hop capacity before policy state changes.
+- Direct TCP plans reject port zero, zero or excessive timeouts, excessive attempts, unapproved IPs, non-canonical IPv4-mapped IPv6 sockets, and IPv6 flow or scope metadata not represented in destination authority before connection I/O.
+- Direct connection code accepts only an explicit `SocketAddr`, never a hostname, and does not read proxy environment variables.
+- Established streams are discarded when peer inspection fails or the observed remote IP or port differs from the approved socket.
+- TLS accepts only an already verified direct stream, never a hostname or new socket, and requires the TLS origin to match the transport-authority origin exactly.
+- DNS TLS identity requires an applicable subjectAltName and never falls back to Common Name; literal IPv4 and IPv6 origins require exact IP subjectAltName entries.
+- TLS uses an explicit immutable trust-root bundle and fixed verification time, and permits only TLS 1.2 and TLS 1.3.
+- TLS trust-bundle policy identifiers must contain at least one ASCII alphanumeric character; punctuation-only labels are rejected while `.`, `_`, `:`, and `-` remain permitted.
+- TLS resumption, 0-RTT, secret extraction, key logging, client certificates, certificate compression, and dangerous custom verifier hooks are disabled in the first slice.
+- The operating-system peer is rechecked before, during, and after the deadline-bound TLS handshake.
+- ALPN selection is restricted to the caller's bounded allow-list, while absence is either explicitly recorded or rejected by policy.
+- Revocation is reported as not configured; the product makes no OCSP or CRL validation claim without supplied revocation evidence.
+- Every generic network header and query value is redacted before evidence leaves the trusted boundary, including conventionally benign field names containing attacker-controlled bytes.
+- Evidence capture enforces count and byte bounds and rejects credential-bearing source URLs, query strings, fragments, controls, whitespace, malformed percent escapes, encoded separators, dot segments, and backslash paths.
+- Network-evidence paths and provenance source URL paths accept only RFC 3986 literal `pchar` syntax plus validated percent-encoded octets and slash separators, preventing raw general delimiters such as `[` and `]` or other invalid URI-presentation bytes from entering either evidence surface.
+- Hard RAM and VRAM pressure pauses the active agent and rejects new admission; hard VRAM pressure also offloads a resident local model.
+- The hourly product agent has no Git metadata or repository authority. A separate post-verification publisher opens one PR and cannot approve or merge it.
+- The unprivileged OpenCode user is restricted to loopback egress during model execution, preventing runner-wide allow-listed endpoints from becoming direct source-exfiltration channels.
+
+[Unreleased]: https://github.com/ContextualWisdomLab/OriginWeave/compare/main...HEAD
diff --git a/docs/evidence/product-technical-gap-baseline-through-2026-09-09.md b/docs/evidence/product-technical-gap-baseline-through-2026-09-09.md
new file mode 100644
index 000000000..c4a87e6c7
--- /dev/null
+++ b/docs/evidence/product-technical-gap-baseline-through-2026-09-09.md
@@ -0,0 +1,1297 @@
+# Product and Technical Gap Baseline
+
+This is a dated delivery baseline, not a substitute for the PRD, TRD, roadmap, architecture decisions, or live GitHub state. It keeps buyer-visible gaps, current issues, active pull-request evidence, and commercial completion tracks in one discoverable place. Protected `main` is the implementation boundary: code in an open pull request is not shipped behavior.
+
+## Current live delivery state
+
+This volatile section is refreshed from live GitHub state and is authoritative only for the exact observations recorded here. The dated snapshot below remains historical evidence and is not promoted to current acceptance evidence. Live GitHub PR/base/head/check APIs are authoritative over PR bodies and prior maintenance prose; a body that still names an older head is stale evidence, not merge evidence.
+
+### Latest verified cut: 2026-09-09
+
+#### Presentation identity and controlled-browser evidence
+
+At `2026-09-09T08:49:58Z`, the live inventory was **131 open pull requests:
+14 Ready/non-draft and 117 Draft; 14 open non-PR issues**. Protected `main`
+remains `87c4daa1830bac5a5228b6036752ad5633232085`; no GitHub Release exists.
+Ruleset `18156473` still requires one counted approval and seven central required
+workflows. Queued reviewer evidence is non-passing: discard queued, skipped,
+cancelled, absent, predecessor, synthetic, status-only, and model-only evidence.
+
+Issue #292 remains the buyer-visible presentation-identity gap. Draft #293 is
+at exact head `6855e2578ae94279cc9ab4a14527b016e8c049ee` on Ready #229
+`024f63690cf05cfe6f0d4a430f0e18ea8fd2c4d6`. Its reusable lifecycle deliberately
+does not apply reduced motion without a complete restore or disposable-context
+owner. Thus command planning cannot be presented as full profile application.
+Draft successor #298 is at exact head
+`d01f45c2c8ac7b0fc4dbc3d3ada60238732cdf8c` on that #293 head. Its non-force
+merge of #305 repaired an older source contract so it now asserts both that
+ReducedMotion remains a discoverable protocol capability and that the reusable
+planner exposes no media-mutation command. This removes an API/documentation
+contradiction only; #298's Draft checks are skipped and it does not establish
+live Chromium transport, page observation, cleanup, protected-main shipment,
+or release acceptance.
+An active controlled evidence harness is implemented, but it does not establish a
+product Browser Session adapter: its exact browser lane failed 0/3 at session
+creation before navigation, so application, page-observed post-conditions, cleanup
+acceptance, and release remain unproven. Issue #212 owns the sandbox-helper contract;
+the product branch must consume it rather than copy or weaken workflow behavior.
+
+#### Transport-closure verification and test-integrity repair
+
+The current queue also leaves Ready #287 at
+`3975daf48e01a5e9d1cf9fb104a3be1aa03b0402` with exact-head native and central
+checks queued. Its prior CodeQL dispatch outcome was an absent central verdict,
+not a classifier-source finding. No release, protected-main shipment, or approval
+is inferred from local work, a child merge, or queued checks.
+
+### Previous verified cut: 2026-09-08
+
+#### Published intent acknowledgment verification
+
+Published #271 head
+[`46db0045904f0289738df843d0a2f179c26673d3`](https://github.com/ContextualWisdomLab/OriginWeave/commit/46db0045904f0289738df843d0a2f179c26673d3)
+retains parent #270 `8eda96915dbbe4cc617f834267c7464689c2844d` and is **not shipped**.
+The previous head's formatter and coverage failures were reproduced. The repair
+retains connection and command pre-consumption guards, removes only a proven
+unreachable duplicate check, and tests genuine socket replies, pending-request
+preservation and private-text-safe diagnostics.
+
+Frozen-source local coverage is **1406/14894/19018/1552**
+(functions/lines/regions/branches), all 100%. Nine focused integration tests,
+148 repository contracts, complete workspace coverage tests, strict Clippy,
+formatting, five doctests and warnings-denied rustdoc pass. Actual Edge visual
+inspection verified the rendered acknowledgment contract and readable layout.
+Hosted [run 34186280263](https://github.com/ContextualWisdomLab/OriginWeave/actions/runs/34186280263)
+is terminal SUCCESS on this exact head: Rust contracts and Production coverage
+both pass. Both review threads are resolved and GitHub reports the Draft head
+**CLEAN**. Protected-parent adoption, governing approval and real browser
+outcome evidence remain required; this is still **not shipped**.
+
+Separately, #255 advanced from superseded test-only head
+`10d5e1ff78d46bbad004d4e0971d0ffceac757fa` to published repair
+`6b6c90ed3919ea84b69527ab688a087eeb45224d`. A real server-sent Close 1010
+first reproduced RED because the client mirrored the client-only code. The
+role guard now rejects it before reply or closure evidence while the paired
+server 1011 control remains valid. Exact-head local coverage is
+**1105/11574/14787/1218**, all 100%; 142 repository contracts, strict Clippy,
+formatting, doctests and warnings-denied rustdoc pass. Actual Edge inspection
+verified the rendered role, masking-key, control-budget and deadline contract.
+Hosted [run 34188785932](https://github.com/ContextualWisdomLab/OriginWeave/actions/runs/34188785932)
+is terminal SUCCESS: Rust contracts and Production coverage both pass on that
+exact head. The Draft is CLEAN, but this remains neither protected-main nor
+shipped behavior.
+
+Draft #293 advanced from visually inspected predecessor
+`6b5241c164f5283f8dd51b1846ef0e4dacec0b29` to exact head
+`0c077445d73640a6299ea4d379faa4b0ab0226c2` on parent #229. Its executed RED showed that standard BiDi was
+overstating complete Screen and Languages support. The minimal repair retains
+only Viewport, DevicePixelRatio, TimeZone and ReducedMotion, pins the immutable
+3 September 2026 dated W3C Working Draft identity, and adds an explicit typed
+viewport/DPR reset intent because session teardown does not clear that override.
+The current head binds three command intents for those four surfaces plus one
+cleanup intent to one bounded opaque browsing context without sending them or
+minting evidence. Exact local coverage is **594/4974/5980/754**, all 100%; 159
+repository contracts, strict Clippy, formatting, doctests and warnings-denied
+rustdoc pass. Exact-head Edge inspection was attempted, but the foreground Edge
+tab was a different repository and was not navigated; current-head visual acceptance is unproven.
+Generated-rustdoc inspection is separately blocked by local-file browser policy.
+The Draft is CLEAN; real Chromium application and page-observed evidence remain
+in #292.
+
+Both source writer leases are released. Inventory remains 126 open PRs, 14
+issues and zero releases. Earlier sections below retain dated evidence, not the
+current status of their superseded runs.
+
+#### Published repeated-control repair
+
+Published #255 head
+[`3a8e6f4f89db2f53c144adb3351c153d89adca58`](https://github.com/ContextualWisdomLab/OriginWeave/commit/3a8e6f4f89db2f53c144adb3351c153d89adca58)
+retains parent #252 `363a78e36e7690e9ed5bf49829567e00e2ec5d59` and is **not shipped**.
+The two-Pong real-peer case first failed. The observer now handles up to **64**
+pre-Close Ping/Pong frames under one total deadline, with fresh caller-supplied keys
+consumed only by Ping responses. A separate key remains reserved for Close.
+Tests prove exact-budget success, **65th** control rejection, exhausted and reused
+key rejection without an extra reply, and literal separately masked Pong payloads.
+
+Frozen-final-source local coverage is **1104/11553/14758/1216**
+(functions/lines/regions/branches), all 100%. All **20 closure** integration tests,
+six controlled-clock tests, 142 repository contracts, strict Clippy, doctests, and
+warnings-denied rustdoc pass. Actual Edge inspection verified the rendered key-slice
+API, local resource limit and total-deadline contract. Hosted
+[run 34184829970](https://github.com/ContextualWisdomLab/OriginWeave/actions/runs/34184829970)
+is queued at this checkpoint: local verification is **not hosted acceptance**.
+
+The repeated-control implementation gap is closed locally, not released. Current-head
+review/check acceptance and protected-parent adoption remain required; **process exit**
+and **profile cleanup** still lack runtime evidence. Inventory remains 126 open PRs,
+14 issues and zero releases. The following same-day sections preserve earlier
+checkpoints; their pending implementation claims are superseded by this section.
+
+#### Published deadline and Close-code repairs
+
+Published #255 head
+[`8716b9d441960a112446c1f89ab417ad9abe28d2`](https://github.com/ContextualWisdomLab/OriginWeave/commit/8716b9d441960a112446c1f89ab417ad9abe28d2)
+retains parent #252 `363a78e36e7690e9ed5bf49829567e00e2ec5d59` and is **not shipped**.
+One operation-wide monotonic deadline now covers the complete closure exchange.
+A delayed real-peer test first reproduced budget renewal; controlled-clock tests
+then verified every deadline transition and final evidence admission.
+
+The shared Close validator also rejects unassigned protocol codes **1016–2999**
+against the current IANA registry, while preserving application/private ranges.
+The boundary test first failed on 1016; real peers now verify no reply or closure
+evidence for 1016, 2000, and 2999. Local functions/lines/regions/branches are
+**1101/11492/14662/1218**, all 100%. All **18 closure** integration tests,
+**142 repository** contracts, strict Clippy, doctests and warnings-denied rustdoc pass.
+Hosted [run 34183766437](https://github.com/ContextualWisdomLab/OriginWeave/actions/runs/34183766437)
+is pending at this checkpoint: local verification is **not hosted acceptance**.
+
+The fresh inventory remains 126 open PRs (12 Ready, all BLOCKED; 114 Draft),
+14 issues and zero releases; protected main remains
+`87c4daa1830bac5a5228b6036752ad5633232085`. The current rule still requires one
+counted approval and seven central workflows. Remaining #255 work is **repeated Ping/Pong**
+with fresh masking keys; browser **process exit** and **profile cleanup** remain unproven.
+Publication still requires protected source, an explicit version and complete release
+artifacts/evidence. The earlier checkpoint below records its own earlier state,
+not the current implementation gaps.
+
+#### Transport-closure verification and test-integrity repair
+
+The latest inventory contains **126 open pull requests: 12 Ready/non-draft and
+114 Draft; 14 open non-PR issues**. All 12 Ready candidates remain BLOCKED.
+Protected main remains `87c4daa1830bac5a5228b6036752ad5633232085`, with
+no package release. These counts do not establish review-thread resolution.
+
+PR #255 has adopted parent #252 at
+`363a78e36e7690e9ed5bf49829567e00e2ec5d59`; the previous instruction to perform
+that adoption is now historical. It remains Draft and is not protected-main acceptance.
+
+- Verified predecessor: [`d126242c7198c447d0fab7983d529441340fd1c9`](https://github.com/ContextualWisdomLab/OriginWeave/commit/d126242c7198c447d0fab7983d529441340fd1c9)
+  passed both Rust contracts and Production coverage in
+  [run 34179452950](https://github.com/ContextualWisdomLab/OriginWeave/actions/runs/34179452950).
+  Full local functions/lines/regions/branches were **1090/11218/14328/1210**, all 100%.
+- Published follow-up: [`07ef43ec71b6dbd8540629bf1df5a63b81541ee4`](https://github.com/ContextualWisdomLab/OriginWeave/commit/07ef43ec71b6dbd8540629bf1df5a63b81541ee4)
+  corrects three masking-rejection test peers. Propagating peer failures first
+  reproduced two false-green fixtures; all 16 closure tests and the full workspace
+  suite including doctests then passed. Hosted
+  [run 34180304951](https://github.com/ContextualWisdomLab/OriginWeave/actions/runs/34180304951)
+  was queued at this checkpoint; predecessor coverage does not prove this head passed.
+
+The direct writer tests compare literal Close bytes with and without a status code,
+invalid deadlines, and adjacent masking-key rejection. The repaired observer fixtures
+consume the real preceding text/Pong before testing reuse and verify no rejected
+response bytes. Broad transport errors with ignored server failures were insufficient.
+
+Remaining buyer-visible gap: repeated Ping/Pong traffic and an operation-wide deadline
+still need implementation and acceptance; unassigned Close status handling needs
+protocol review. Browser-process exit and profile cleanup remain unproven. The next
+source work belongs to #255 before downstream adoption; no PR is closed as a substitute
+for carrying its delta. Actual Edge inspection verified the PR page and corrected
+evidence presentation, not the OriginWeave product browser. Package publication remains
+withheld until protected-source, version, artifact and release requirements are met.
+
+### Historical verified cut: 2026-09-07
+
+#### Published session-end reply binding: 05:35 UTC
+
+The refreshed five-page inventory contains **125 open pull requests: 12 Ready/non-draft
+and 113 Draft; 14 open non-PR issues**, with zero unresolved review threads and no
+pagination gaps. All Ready candidates remain BLOCKED. Protected main remains
+`87c4daa1830bac5a5228b6036752ad5633232085`; no tags or releases exist. Active ruleset
+`18156473` requires one counted approval and seven central workflows; the author
+remains the only listed collaborator. This reviewer-provisioning gap permits no bypass.
+
+Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
+
+| Owner | Publication | Exact head | Parent | Local coverage (F/L/R/B) |
+| --- | --- | --- | --- | --- |
+| #251 | Published; Draft | [`924ad975`](https://github.com/ContextualWisdomLab/OriginWeave/commit/924ad97551750d4a901ded38b89488cc5438e54f) | #250 `bbdc6ace` | 1057/10840/13873/1194 |
+| #252 | Published; Draft | [`363a78e3`](https://github.com/ContextualWisdomLab/OriginWeave/commit/363a78e36e7690e9ed5bf49829567e00e2ec5d59) | #251 `924ad975` | 1064/10898/13947/1194 |
+
+Both heads passed all four coverage dimensions at 100%, 142 Python contracts,
+compileall and full stable Rust formatting, check, tests, strict Clippy and
+dependency-inclusive warnings-denied rustdoc. Actual Edge inspection covered the
+rendered APIs, the end-response link to its sealed-reply type, and published commits.
+Independent source review found no actionable findings; it is not counted approval.
+
+#251 preserves its sender and original tests while adopting status-reply safeguards.
+#252 reproduced replacement end-reply acceptance before and after parent adoption,
+then reused connection-bound registration and sealed reply correlation. Replacement
+success and error replies now leave the original request pending, and its original
+reply can still complete it. The fixtures use retained receipts after separate
+listener servers finish: they do not establish same-endpoint or live-stream recovery,
+browser shutdown, profile cleanup or protected delivery.
+
+Coverage artifact SHA-256 for #251 is
+`5cbb52b50b5740a22d269490f6296380d4b53bbe61ceae632376f6de63ff1054`;
+for #252 it is `9f2249637f31916acf9874baec724ddedfe621e5b14fdb6f4bd850620e89f307`.
+Exact-head CI `34085877650` (Rust `101629699915`, coverage `101629699750`) and
+`34087239755` (Rust `101633511589`, coverage `101633511320`) are queued.
+Local verification is not hosted acceptance. Both source writer leases are released.
+
+Next, safe successor #255 must adopt #252 while preserving its stronger closure
+provenance; #253/#254 remain open and are not safe intermediate prerequisites.
+New issue #292 owns version-pinned browser presentation evidence after #229; its
+requested implementation and real-browser acceptance remain unverified. Existing
+release, runtime, extraction, API and enterprise gaps remain open. Automatic package
+publishing is still unattempted: the protected package disables publication, no
+Python package or registry-publishing workflow exists, and release evidence plus an
+explicit version decision are required. Secret metadata is not release readiness;
+organization key availability remains unverified. No gates or credentials changed.
+Earlier checkpoints remain unchanged historical evidence.
+
+#### Published status-response repair: 04:45 UTC
+
+The refreshed five-page inventory contains **125 open pull requests: 12 Ready/non-draft
+and 113 Draft; 13 open non-PR issues**, with zero unresolved review threads and no
+pagination gaps. All Ready candidates remain BLOCKED. Protected main remains
+`87c4daa1830bac5a5228b6036752ad5633232085`; no tags or releases exist. Active ruleset
+`18156473` requires one counted approval and seven central workflows; the author is
+the only listed collaborator. This reviewer-provisioning gap does not permit bypass.
+
+Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
+
+| Owner | Publication | Exact head | Parent | Local coverage (F/L/R/B) |
+| --- | --- | --- | --- | --- |
+| #250 | Published; Draft | [`bbdc6ace`](https://github.com/ContextualWisdomLab/OriginWeave/commit/bbdc6ace7a5932adf24836700f806850e6b230bc) | #249 `65ac3ab9` | 1043/10711/13717/1188 |
+
+All four coverage dimensions are 100% locally. Prior exact-head CI `34064507212`
+failed formatting and coverage; that terminal evidence supersedes its historical
+queued description. The repair reuses the sealed-reader tests and a shared, pure
+reply-routing check. Validation and connection provenance still precede completion.
+Public loopback checks reject events, unattributable errors and unbound requests
+without consuming pending work. After an unbound request rejects a sealed reply,
+the original request can still accept it. This does not prove a browser outcome.
+
+All 142 Python contracts, compileall and full Rust formatting, check, tests, Clippy
+and warnings-denied workspace documentation passed at the published head. Coverage
+artifact SHA-256: `e1fccefc6b56eabe653ae41377fbcec0aa89b4be1ada92dba73d3ef87e841029`.
+Actual Edge inspection confirmed readable API documentation and the published commit.
+Independent source review found no actionable findings; it is not counted approval.
+
+Exact-head CI `34084134654` has Rust job `101624834145` and coverage job `101624834026`
+queued. Local verification is not hosted acceptance. The source writer lease is
+released; this documentation checkpoint needs its own verification and visual inspection.
+Earlier checkpoints remain unchanged historical records, not current acceptance.
+
+Automatic publishing remains unattempted: the protected package disables publication,
+no Python package or publishing workflow is present, and release evidence and an
+explicit version decision are still required. Empty repository/environment secret
+metadata does not prove organization keys absent; their listing remains unavailable
+under current permissions. No credential values, workflow gates or release settings
+were changed. Next executable work remains parent-first status-receipt adoption,
+protected foundations and release-readiness evidence under their separate owners.
+
+#### Published semantic-action adoption: 04:11 UTC
+
+The five-page inventory contains **125 open pull requests: 12 Ready/non-draft
+and 113 Draft; 13 open non-PR issues**, with zero unresolved review threads and no
+pagination gaps. All Ready candidates remain BLOCKED. Protected main remains
+`87c4daa1830bac5a5228b6036752ad5633232085`; no tags or releases exist.
+Active ruleset `18156473` still requires one counted approval and seven central
+workflows; the only listed collaborator is the author. Advisory review is not approval.
+
+Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
+
+| Owner | Publication | Exact head | Parent | Local coverage (F/L/R/B) |
+| --- | --- | --- | --- | --- |
+| #93 | Published; Draft | [`82056d13`](https://github.com/ContextualWisdomLab/OriginWeave/commit/82056d13aa94c106060b84ee76be56fcb7787fc8) | #271 `b0410ae9` | 1403/14852/18976/1552 |
+| #95 | Published; Draft | [`6b29d890`](https://github.com/ContextualWisdomLab/OriginWeave/commit/6b29d890245ed2f612c2998198f4e8c8a06da312) | #93 `82056d13` | 1408/14897/19022/1552 |
+| #96 | Published; Draft | [`cbabf55c`](https://github.com/ContextualWisdomLab/OriginWeave/commit/cbabf55c6a25b979fa0d9e3c1677338665975ab7) | #95 `6b29d890` | 1410/14908/19030/1552 |
+
+Each listed coverage dimension is 100% locally. These ordinary parent adoptions retain
+node-action binding, explicit policy allow and dispatch-time node validation while
+inheriting connection-bound reply rejection. The original connection can still complete
+its own pending request; unrelated pending work survives. The child implementations
+and original tests were preserved. Node authority, policy decisions, callback execution
+and reply provenance remain separate checks, not proof of the requested browser outcome.
+
+#93 passed 10 focused Rust tests and 149 Python contracts; #95 passed eight focused tests
+and 150 Python contracts; #96 passed six focused tests and 151 Python contracts. Complete
+local Rust gates passed. #93's final Python/Markdown-only correction preserved the entire
+measured Rust tree and manifests. Coverage artifact SHA-256 values, in owner order:
+`9479014b94c1575a379d31fea09309deaf00143374b6dd6f04a539aaf452a685`,
+`666797dec8486f0f196616525303bd24dc52ef5d035f1c548e1bddde3fb48a22`,
+`f69829e1968a75c8d3c8ea7d08df87669d47a7d6adfbbb88d08d9bb0d384b10d`.
+
+Exact-head CI `34079739018`, `34080772063` and `34081979602` remain queued.
+#93's MV3 run `34079739021` is also queued. Independent read-only review found no
+remaining actionable findings; it is not counted approval. Source leases are released.
+All three revisions received actual visual inspection in isolated Edge, with readable
+API documentation and no observed clipping or overlap. Dependency-inclusive workspace
+rustdoc restored cross-crate links omitted by the no-dependencies build. This is
+documentation presentation, not product-browser acceptance.
+
+The user-requested `AGENTS.md` lessons were published at `de06d65e` with 197 Python
+contracts and their own rendered visual inspection. Automatic package publishing was
+checked but not attempted: protected main's core manifest has `publish = false`, no
+publishing workflow or Python package manifest is present, and the release contract
+still requires its evidence and an explicit version decision. Repository/environment
+secret metadata lists were empty; organization-secret listing returned 403, so those
+keys remain unverified, not proven absent. No secret values were accessed.
+
+Next are status receipts, protected-foundation work and release-readiness evidence.
+#195/#279 workflow recovery and #212 sandbox acceptance retain their separate owners.
+This baseline requires its own exact-head verification and visual inspection; it cannot
+inherit source coverage or hosted acceptance. Earlier checkpoints below are unchanged
+historical observations. No merge, package publication or release is claimed.
+
+#### Published observation safeguards: 03:00 UTC
+
+The complete five-page inventory contains **125 open pull requests: 12 Ready/non-draft
+and 113 Draft; 13 open non-PR issues**, with zero unresolved review threads and no
+pagination gaps. All Ready candidates remain BLOCKED. Protected main remains
+`87c4daa1830bac5a5228b6036752ad5633232085`; no tags or releases exist.
+Active ruleset `18156473` requires one counted approval and seven central workflows;
+the sole collaborator is the author. Advisory review does not satisfy that approval.
+
+Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
+
+| Owner | Publication | Exact head | Parent | Local coverage (F/L/R/B) |
+| --- | --- | --- | --- | --- |
+| #270 | Published; Draft | [`8eda9691`](https://github.com/ContextualWisdomLab/OriginWeave/commit/8eda96915dbbe4cc617f834267c7464689c2844d) | #269 `3df2a631` | 1346/14113/17928/1464 |
+| #271 | Published; Draft | [`b0410ae9`](https://github.com/ContextualWisdomLab/OriginWeave/commit/b0410ae92bd20eaf31d09b7d49390e13cb045999) | #270 `8eda9691` | 1393/14770/18899/1546 |
+
+Every listed coverage dimension is 100% locally. #270 repaired wrong-session dispatch,
+invalid deadlines and reused masking keys through existing transport guards, preserving
+the fixed observation command and ordinary parent history. Its 11 focused tests,
+147 Python contracts and complete stable Rust gates passed. Coverage artifact SHA-256:
+`7748d42c541140314995ffa2087b3aca87d52875f98a6ffa36614a0453b30260`.
+
+#271 reproduced replacement-connection completion at RED `29cd0d66`, still failing after
+parent adoption `14efb678`. The existing sealed receipt and connection-aware correlation
+now reject foreign success, protocol error and script exception. A same-listener test
+proves the original connection can still complete its command, leaving unrelated work
+pending. Earlier fixture compile errors are not runtime RED evidence. All 148 Python
+contracts, focused recovery tests and complete stable Rust gates passed. Coverage SHA-256:
+`8f211170f9e7b691b95b12151e9e87827e4cb741ca8fa2892576adf8421e8a30`.
+
+Exact-head native CI `34076117534` and `34077987302` remain queued, not passing evidence.
+Independent read-only review found no remaining actionable findings; it is not counted
+approval. Source leases `5563898819` and `5564211278` are released. Both revisions received
+actual visual inspection in isolated Edge: generated API documentation and published
+commit views were readable without observed clipping or overlap. This is documentation
+presentation, not product-browser acceptance. This baseline needs its own verification
+and visual inspection; it cannot inherit source coverage or hosted acceptance.
+
+Next are downstream semantic-action adoption, status receipts and protected-foundation
+work. Matching text replies do not establish authenticated browser ownership, policy
+approval or complete authorized action success. #195/#279 workflow recovery and #212
+sandbox acceptance retain their separate owners. Earlier checkpoints below are unchanged
+historical observations, including superseded next actions; no release is claimed.
+
+#### Published response and observation: 01:45 UTC
+
+The fresh five-page inventory still contains **125 open pull requests: 12 Ready/non-draft
+and 113 Draft; 13 open non-PR issues**, with zero unresolved review threads.
+All Ready candidates remain BLOCKED.
+Main remains `87c4daa1830bac5a5228b6036752ad5633232085`; there are no tags or releases.
+Active ruleset `18156473` requires one counted approval and seven central workflows;
+the sole collaborator is the author. No advisory review is counted as that approval.
+
+Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
+
+| Owner | Publication | Exact head | Parent | Local coverage (F/L/R/B) |
+| --- | --- | --- | --- | --- |
+| #268 | Published; Draft | [`ff27220c`](https://github.com/ContextualWisdomLab/OriginWeave/commit/ff27220cb5eb4d11ca1dc5614a4181e1a397a3f1) | #267 `ebd507ae` | 1325/13907/17682/1456 |
+| #269 | Published; Draft | [`3df2a631`](https://github.com/ContextualWisdomLab/OriginWeave/commit/3df2a631bacd7109b3982fdd7ac599d0bd92a589) | #268 `ff27220c` | 1338/14025/17843/1460 |
+
+Each dimension is 100% covered locally. #268's actual RED `716fd842` reproduced three
+pointer/session failures before ordinary parent adoption. The sealed text consumer and
+all original text tests were preserved. Review exposed an unqualified historical limitation;
+RED `49d18f5f` reproduced it, and the correction anchors it to predecessor `35cb1197`.
+#269's RED `658fb676` reproduced dispatch to the wrong browser session before ordinary
+merge `f883fd6f`. Its fixed field-observation command and all eight child tests remain
+byte-identical to the predecessor; the entire network crate matches its parent.
+
+#268 passed 20 focused tests and 146 Python contracts; #269 passed 18 focused tests and
+147 Python contracts. Both passed complete local Rust 1.97.1 gates, compileall, CodeGraph
+and diff checks. Coverage artifact SHA-256 values are respectively
+`bce553c93d5b5dd3cb59c78bbb75c9f1225b659ec7ea22b674c54455e5067a07` and
+`1e203e6c0801bedbbcdbeb1be500b533a3d68e0ac842de221f7f470c5e742c66`.
+The pinned nightly branch-instrumentation warning remains visible.
+
+Exact-head native CI `34072645796` and `34073733364`, and #269 MV3 `34073733355`,
+are queued, not passing evidence. Independent read-only reviews found no remaining
+actionable findings. Source leases `5563501468` and `5563726066` are released.
+Both published PRs and dated dossiers received actual visual inspection in isolated Edge:
+revision links, current/historical separation and text were readable without visible
+clipping or overlap. This is GitHub documentation presentation, not product-browser acceptance.
+This baseline requires its own exact-head tests and visual inspection.
+
+Next is #270's content-aware adoption of #269, followed by response/value verification,
+status receipts and protected-foundation work. Request construction and protocol replies
+still do not prove field changes, authenticated browser ownership, policy approval or
+authorized action success. #195/#279 workflow recovery, #212 sandbox acceptance and
+.github#1929 dispatch identity admission retain their separate owners. Earlier checkpoints
+below remain unchanged historical observations, including their superseded next actions.
+
+#### Published input descendants: 00:40 UTC
+
+The latest complete five-page review/check/thread inventory contains **125 open pull requests:
+12 Ready/non-draft and 113 Draft; 13 open non-PR issues**, with zero unresolved review threads
+and no unfinished pagination. All 12 Ready candidates remain BLOCKED. Current rules require
+one counted approval and seven required workflows; the sole collaborator is the author.
+Protected main remains `87c4daa1830bac5a5228b6036752ad5633232085`, with no tags or releases.
+REST release inventory was rate-limited; GraphQL independently confirmed both empty inventories.
+
+Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
+
+| Owner | Publication | Exact head | Parent | Local coverage (F/L/R/B) |
+| --- | --- | --- | --- | --- |
+| #265 | Published; Draft | [`e94a2372`](https://github.com/ContextualWisdomLab/OriginWeave/commit/e94a2372fe3771f9ddf70291d34fc9a7e4770ec9) | #264 `43395711` | 1297/13618/17315/1442 |
+| #266 | Published; Draft | [`e3885f69`](https://github.com/ContextualWisdomLab/OriginWeave/commit/e3885f69df2cf3899184209efdee5b11bba1bd86) | #265 `e94a2372` | 1310/13766/17524/1452 |
+| #267 | Published; Draft | [`ebd507ae`](https://github.com/ContextualWisdomLab/OriginWeave/commit/ebd507ae56c3064e3cae5566502f539c20618a8f) | #266 `e3885f69` | 1318/13852/17610/1456 |
+
+Functions, lines, regions and branches are each 100% covered on these exact local source trees.
+#265 replay `e7fb1527` first failed both replacement-reply tests before ordinary merge
+`d847b530`. The contributor's existing outbound-session guard remains intact. #266 replay
+`009f9a41` and #267 replay `4e020e16` each failed all three inherited pointer regressions
+before ordinary merges `d9503b30` and `7e4bd76d`. The repaired click path rejects a foreign
+session before reserving or sending work, and rejects replacement success/error replies
+without consuming either pending request. The original reply completes only its own request.
+
+#265 passed 14 focused tests; #266 passed 21; #267 passed 23 focused tests. Each exact head
+passed full local Rust 1.97.1 gates, 145 Python contracts, compileall, CodeGraph and diff checks.
+The text constructor, privacy checks and eight core text tests were preserved in #266.
+#267 also preserves its existing text sender, public exports and ten text-send tests unchanged.
+Coverage artifact SHA-256 values for #265, #266 and #267, respectively, are
+`fc9aebce5e633be1c4faa10755e56bb792a0f2679daeef3a0a1984d4a4b58404`,
+`a06aebb7c2f30c153c3a3867f3b7847d9c485ae2db5ba2b23029ca0375e4ec79` and
+`8bd32ad83ba5f9ef65bad8a3503f5e6cff38335b8526bff9f974bd6ee7fabceb`.
+The pinned nightly branch-instrumentation warning remains explicit; no coverage gate changed.
+
+Native CI `34068277243`, `34068882527` and `34070121583` remains queued for those three heads;
+#265/#266 MV3 runs `34068277244` and `34068882439` are also queued. Independent read-only
+preservation reviews found no actionable findings, but are not counted approval. Source writer
+leases `5563064004`, `5563186050` and `5563277731` are released. Actual visual inspection
+used actual screenshots of published exact-head evidence: in-app for #265/#266 and isolated
+installed Edge for #267 after browser-control connection failures. Revisions, caveats and
+wrapping were readable without observed clipping or overlap. This is GitHub presentation,
+not product-browser acceptance. This baseline needs its own checks and visual inspection.
+
+Next source work is #268's content-aware adoption of #267, retaining its existing sealed
+text-response consumer rather than rebuilding it. Then continue status receipt prerequisites
+and protected-main asset preservation at their canonical owners. #195/#279 workflow recovery,
+#212 sandboxed browser acceptance and .github#1929 dispatch identity admission remain separate.
+Text transport and pointer acknowledgments do not prove field values, browser authentication,
+policy approval, navigation causality or release acceptance. Earlier dated observations below
+remain unchanged historical evidence and do not override this checkpoint.
+
+### Latest verified cut: 2026-09-06
+
+#### Published subscription descendants: 23:27 UTC
+
+The complete five-page live inventory contains **125 open pull requests: 12 Ready/non-draft and
+113 Draft; 13 open non-PR issues**. It has zero unresolved review threads and no
+unfinished review/check/thread pagination. This supersedes the older sole-#147
+thread observation below. All 12 Ready candidates remain BLOCKED. The active ruleset
+still requires one counted approval and seven required workflows; only one repository
+collaborator is available, so independent reviewer provisioning remains unresolved.
+Protected main remains `87c4daa1830bac5a5228b6036752ad5633232085`; there are no tags or releases.
+
+| Owner | Publication | Exact head | Parent | Local coverage (F/L/R/B) |
+| --- | --- | --- | --- | --- |
+| #263 | Published; Draft | [`4868d3e9`](https://github.com/ContextualWisdomLab/OriginWeave/commit/4868d3e9f19133ac3382ee8532878aef27468893) | #277 `46ae62aa` | 1244/13053/16729/1422 |
+| #264 | Published; Draft | [`43395711`](https://github.com/ContextualWisdomLab/OriginWeave/commit/433957117ad9e29b26715b062f5adcc9789744ba) | #263 `4868d3e9` | 1282/13461/17140/1440 |
+
+F/L/R/B denotes functions, lines, regions and branches; every metric is 100% covered.
+
+Both owners adopted their parents by ordinary merges after actual replacement-reply
+regressions failed. #263 replay `90395f81` failed 0/2, followed by merge `9e85cadc`
+and sealed-reader fixture adaptation `cb0c4261`. Its 10 focused tests and full local
+Rust 1.97.1 gates passed, alongside 144 Python contracts. #264 replay `637fd97d`
+failed 0/2 before merge `92e576c8`; its 13 focused tests, full local Rust 1.97.1 gates
+and 145 Python contracts passed. Both passed compileall, CodeGraph and diff checks.
+The unchanged pinned coverage verifier passed the counts above; experimental branch
+instrumentation remains explicit. Their coverage artifacts have SHA-256 values
+`1e51cd7d08ebacc80ecadb568e0bca94384ffc0e1d721237b018cae31daa1336` and
+`e102bda45da886bc3138b981dac73f4239e2f6e5cd893688f10ed72b1f5af771`, respectively.
+
+The child keeps its original browser-state association, increasing request identifiers,
+separate raw/typed traffic and consuming subscription-shutdown ownership. Original
+replies still complete their own request after a foreign reply is rejected; unrelated
+work remains pending. Independent read-only preservation reviews found no actionable
+findings, but are not counted GitHub approval. Writers `5562611932` and `5562770321`
+are released. CI `34065055213` for #263 and CI `34066516991` plus MV3 `34066516992`
+for #264 were queued at the recorded readbacks; predecessor success is not new-head proof.
+
+Actual visual inspection used actual in-app screenshots of both published exact-head
+traceability sections: wrapping and historical/current boundaries were readable without
+observed clipping or overlap. This is GitHub presentation evidence, not product-browser acceptance.
+Earlier locked-Mac and pending-visual observations remain historical. This new baseline
+revision needs its own checks and visual inspection; source coverage does not validate it.
+
+Central dispatch identity admission remains separately owned by .github#1929
+([verified diagnostics](https://github.com/ContextualWisdomLab/.github/issues/1929#issuecomment-5562597667));
+the documented actor/sender mismatch is not evidence of recovery. Do not change owner
+settings or repeat a disproven dispatch route under this documentation task. Next safe
+source work is content-aware descendant adoption, pointer outbound authority and status
+receipt prerequisites, alongside protected-main asset preservation. Browser authentication,
+navigation-event authenticity, causal page effects and release acceptance remain unproven.
+
+#### Connection-bound text responses: 14:54 UTC
+
+This checkpoint supersedes the source status and executable actions in the earlier
+timed checkpoints below, which remain historical evidence. The fresh complete
+five-page inventory still contains **125 open pull requests: 12 Ready/non-draft and
+113 Draft; 13 open non-PR issues**. All Ready candidates remain BLOCKED; only #147 retains unresolved
+thread `PRRT_kwDOTulPlM6coZwc`. Protected main is unchanged at
+`87c4daa1830bac5a5228b6036752ad5633232085`; release and tag inventories are empty.
+
+Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
+
+Published sender #267 `3346d8ecc72932b98ec495d9cc52d6e5727c3064` remains on #266
+`eb6c236ff2f4a58b807a2f2c914bd1ddb6079fb3`. Published response owner
+#268 `e567af9e678fd4791776df795e89ed666975e6c2` adopts it by ordinary merge
+`e1188c86`. Actual socket RED `4632f2df` accepted a reply from a replacement
+connection using the same listener/session. Sealed consumer `d6889c80` exposed
+missing sender provenance; the parent now records its private connection generation
+before I/O and the consumer checks the sealed receipt before consuming pending work.
+Foreign success/error replies retain both pending entries; the original reply still
+completes its request and leaves unrelated work pending. Seven focused response tests
+preserve real remote-error consumption, extensible success, malformed/unknown
+retention, family isolation and missing-provenance rejection.
+
+Both exact source heads pass full local Rust1.97.1 gates, 145 Python contracts,
+compileall, CodeGraph and diff checks. Their function/line/region/branch coverage is
+1317/13835/17598/1456 for #267 and 1324/13890/17670/1456 for #268, each 100%.
+Coverage SHA-256 values are respectively
+`8d7a3687557d067ee99fd512936077c0c9662d40d9d6544232bb5a377dc88e23` and
+`5865d6e69dc5584bad8ef4e866970eb6aa845ad692aa5421a0ffb6f6f01e1540`.
+CI `34040202356` and `34040306372` remain queued; formal reviews are empty.
+Actual Edge screenshots verified the published source PRs' exact heads, Draft state
+and readable evidence without observed clipping or overlap. This is GitHub
+presentation evidence, not product-browser acceptance or protected-main delivery.
+Writers `5559965531` and `5559948553` are released. This documentation revision
+requires its own publication, visual inspection and exact-head checks.
+
+At the 14:54 checkpoint, next source work was to address pointer and status receipt provenance in their existing
+owners, plus the separate pointer outbound session-binding gap, before promoting
+their evidence. Reuse the sealed reader and canonical connection checks; status
+projection must still validate before correlation consumption. Browser ownership,
+policy approval and observed field/DOM postconditions remain unfinished. Current
+foundation #195 `63997bcf555e2c5c8e91ba287734ffba3837a1b7` remains on
+`6922dd98779e8f8aad132a3b1f563d7ba6e6d070`; #279 workflow generation and #212
+sandbox boundaries remain prerequisites. Counted approval and central workflow
+requirements must not be bypassed. Source and documentation writers stay separate.
+
+##### Pointer receipt follow-up: 15:28 UTC
+
+This later checkpoint supersedes the preceding pointer next-action claim, not the
+dated text-entry evidence. Published sender #257 `9451fd8a23dec95b31749376bc78c2eaca977fe8`
+remains based on #256 `881c7f09ee9161ce8664dd75226938ecf60b85e5`; published
+response owner #258 `5417ce32ed957aa166807f1023647caccc2920cb` adopts that sender
+by ordinary merge `0234b587`, preserving the existing hourly writer's test history.
+
+Executed RED `8193fcd5` accepted a replacement connection's reply. Stronger RED
+`d9396f05` held the listener address and session constant and reproduced both foreign
+success and error consumption. Consumer-only `588fe731` failed original-response
+acceptance with missing sender provenance. The integrated sender registers its
+private connection generation before I/O; the consumer validates the existing sealed
+receipt before consuming pending work. Six focused tests pass, including exact
+foreign-connection mismatch, unchanged pending work and original-reply recovery
+with counts **2 → 2 → 1**, extensible success, real remote error, malformed input
+and unknown-id retention. No public receipt constructor or raw fallback was added.
+
+Both source heads passed full local Rust 1.97.1 gates, 142 Python contracts,
+compileall, CodeGraph and diff checks. Exact function/line/region/branch coverage is
+1093/11134/14204/1214 for #257 and 1100/11191/14276/1214 for #258, each 100%.
+Their artifact SHA-256 values are
+`a976c273d86d43eef366044cca5b73566064844612a35ba02ac4b7fe59a0969f` and
+`ed46b5881d952723eec724a74e6ebb8856a0801e4db0570c9f8a6e6cf04eadb5`.
+CI `34041977863` and `34042223733` remain queued; successful advisory statuses
+are not counted approval. Actual Edge screenshots verified both published PR
+heads, parent boundaries and readable evidence without observed clipping/overlap.
+This is GitHub presentation evidence, not product-browser acceptance. Writers
+`5560144371` and `5560167022` are released. This baseline update requires its own
+publication, rendered inspection and hosted checks; source coverage is not its proof.
+
+The next source queue is content-aware descendant adoption, separate pointer
+outbound session authority, and status receipt provenance. Existing status owners
+#249 `84b9407978ae0f6c115f01170b6069c601b21104` and
+#250 `7e85a7e5f0147f4b712129cd19aaa3d0a0a54634` precede the shared sealed-receipt
+and connection-registration helpers: establish the canonical prerequisites in
+dependency order rather than copying a later descendant or creating a cyclic stack.
+Status projection must remain validated before state consumption. Browser ownership,
+policy approval, same-connection freshness and observed page effects remain unproven;
+protected-main approval, central workflows and release gates still apply.
+
+##### Receipt descendant adoption: 16:17 UTC
+
+This checkpoint supersedes the earlier descendant-adoption next action for the
+following four owners. They now include the pointer receipt repair by ordinary
+parent merges; no history was rewritten and no protected merge occurred.
+
+| Owner | Published exact head | Parent | Fresh CI |
+| --- | --- | --- | --- |
+| #259 | `180c168ecbdcd5eb7a4ad14ab4a53e8670646bf7` | #258 `5417ce32` | `34043222194` queued |
+| #260 | `3807aabeb22f9622610c3c8d504d1c686d25d896` | #259 `180c168e` | `34043664230` queued |
+| #261 | `ba100fbc39e1ac4f10ee4faade38418551bb8298` | #260 `3807aabe` | `34044193429` queued |
+| #277 | `121578a43adda12221a9ca9ab8ada0a4fd03efef` | #261 `ba100fbc` | `34044857929` queued |
+
+Canonical regression replays `6999252e`, `8f6b5a0c`, `c03e0dcc` and `3734a874`
+each reproduced a replacement connection consuming an original pending pointer
+command. #259's merge also exposed a raw-response fixture compile failure;
+`d1fd06bf` adopted the existing sealed reader and retained its returned connection
+for the subsequent event. Descendants preserve that repair and the stronger
+foreign-success/error rejection with original-response recovery. Their respective
+focused suites passed 16, 18, 21 and 23 tests; full local Rust 1.97.1 gates passed.
+#259–#261 passed 142 Python contracts each; #277 passed 144 Python contracts.
+
+Exact numerical function/line/region/branch coverage is
+1140/11847/15133/1332, 1154/11975/15318/1334, 1173/12110/15515/1334 and
+1222/12821/16447/1418 respectively, each 100%. The unstable branch-instrumentation
+warning remains visible. #259's Manifest V3 run `34043222184` is also queued.
+Independent read-only reviews found no actionable preservation findings; they are
+not counted GitHub approvals. Actual Edge screenshots verified the published PR
+evidence, not product-browser acceptance. All four source writers are released.
+
+Document-epoch, origin-validation and subscription-specific production/tests remain
+unchanged. Subscription deadlines still fail before registration; proven no-byte
+failures retire only that request, while uncertain writes preserve pending state.
+The next bounded source work is subscription response provenance in #277, followed
+by content-aware descendant integration. Preserve result projection before state
+consumption. Status receipt prerequisites and pointer outbound session authority
+remain separate gaps. The protected-main asset-preservation finding also remains:
+the active source stack lacks this baseline file, so pairwise CI cannot prove safe
+main integration. Browser ownership, event provenance, causality and release gates
+remain unfulfilled. This documentation update requires its own checks and visual
+inspection; source coverage does not validate the documentation revision.
+
+##### Subscription response repair: 16:37 UTC
+
+This checkpoint supersedes the subscription response provenance next action above.
+#277 is published at `46ae62aa31e35c702cd61c16322d05c7a9c35da1`, with unchanged
+#261 parent `ba100fbc39e1ac4f10ee4faade38418551bb8298`. On baseline `122ca139`,
+both success and error replies from a second connection to the same listener/session
+consumed the original subscription (0/2 regressions passed). Repair `8b1508c8`
+reuses private sender registration and the sealed receiving-message consumer;
+replacement replies now leave both requests outstanding and the original reply
+completes only its own request. Required result projection still precedes state
+consumption; deadline and proven no-write safeguards remain unchanged.
+
+All 14 focused subscription tests, full Rust 1.97.1 gates, 144 Python contracts,
+compileall, CodeGraph and diff checks pass locally. Numerical production coverage
+is 1222/12824/16453/1418 functions/lines/regions/branches, each 100%; artifact SHA-256
+is `68183d40d92663f604db48342c0eee0428e18836ed008fee1268e6cf14b5d60a`.
+The unstable branch-measurement warning remains. Exact-head CI `34045953423` is
+queued. Independent source and documentation reviews found no actionable findings;
+they are not counted approval. Source writer `5560556181` is released.
+
+Current-head visual inspection remains pending: the actual browser tool reports
+the Mac is locked and automatic unlock failed. Earlier screenshots do not validate
+this repair or the present baseline update, and are not product-browser acceptance.
+The next safe source queue is canonical outbound session binding, same-connection
+freshness, status receipt prerequisites and protected-main asset preservation,
+with owner-first dependency integration rather than a cyclic descendant merge.
+Navigation-event authenticity and causal page effects remain unproven. No workflow,
+approval, protected merge, tag or release changed. This baseline revision requires
+its own documentation tests, hosted checks and visual inspection.
+
+#### Session repair and child adoption: 13:35 UTC
+
+This update supersedes the 12:28 source/adoption and visual-boundary claims below.
+The complete five-page live inventory read at 13:29 UTC still found **125 open pull requests: 12 Ready/non-draft and 113 Draft; 13 open non-PR issues**.
+All 12 Ready PRs were BLOCKED, with only #147's unresolved thread
+`PRRT_kwDOTulPlM6coZwc`. Subsequent REST reads confirmed 13 non-PR issues, no releases
+or tags, and unchanged protected main `87c4daa1830bac5a5228b6036752ad5633232085`.
+
+Parent #264 is now `6f331a5b220349a1aaa1b1841d5e8ec027b9ad49`, Draft on #263
+`3f22de94b63da83eaa8b5b1270912b21a3ecd006`. Its original actual-socket regression
+was executed and reproduced foreign-session dispatch before the canonical session
+mapping guard repaired it. Unknown, retired and mismatched mappings reject without
+creating state, registering correlation or emitting command bytes. The fixture now
+waits for seed-Pong setup before simulating disconnection; the diagnostic unit-copy
+coverage case is retained. Full Rust 1.97.1 gates, 145 Python contracts and exact
+1282/13455/17136/1440 production functions/lines/regions/branches each pass at 100%.
+Artifact SHA-256 is `b85acd79c3e3a3a49561b9148b6aafadc26cd292a542ec74ef2b523e88c42eaa`.
+Current CI `34035628391` and MV3 `34035628337` remain queued, not hosted GREEN.
+The source writer released in comment `5559331330`; actual Edge visual inspection
+of its published body and Checks page was completed on this head.
+
+Child #265 now publishes `7147893c96ca95c9b5b275d8011c5bfe99aab065` and actually
+adopts that parent. Original parent regression replay `98621adf` first reproduced
+the same socket failure on the child (0 passed, 1 failed). An ordinary merge then
+preserved both histories, admitted-node/current-document authority, typed dispatch,
+deadline and ambiguous-write safeguards, all four previously restored postcondition
+tests, and both Proposed ADR refinements. All 16 focused regressions, full Rust
+1.97.1 gates, 145 Python contracts, compileall, CodeGraph and diff checks pass.
+Exact pinned-nightly coverage is 1296/13601/17303/1442, each 100%; artifact SHA-256
+is `480d6d3ca492565e8a04bfce4b0c135dffbff276532d129c410674838e8fec2f`.
+CI `34036335342` and MV3 `34036335505` remain queued. Fresh actual Edge visual inspection
+verified the published body and Checks page at this head: readable wrapping, no
+observed clipping or overlap, and exact job IDs matching the API. This is GitHub
+presentation evidence, not product-browser acceptance. Writer `5559497861` released
+after normal publication and readback. Independent preservation review found no
+actionable finding; it is not counted approval.
+
+These are local parent/child acceptance results, not protected-main delivery.
+The branch-instrumentation warning, #195/#279 foundation prerequisite, counted-review
+rule, browser authentication, action authorization, pointer-reply connection binding,
+navigation causality and release gates remain separate. Earlier revision measurements
+and screenshots are historical. This new documentation revision still requires its
+own post-publication visual inspection; prior #238 `112ba13a` inspection is not transferred.
+
+#### Text-input adoption and review correction: 13:52 UTC
+
+Published #266 `eb6c236ff2f4a58b807a2f2c914bd1ddb6079fb3` adopts #265 `7147893c`
+through ordinary merge `5a722867`, preserving original child `cc9980c0` and its unchanged
+text-input source/eight tests. Actual inherited socket RED `18a64573` failed before
+adoption. All 24 focused regressions, complete Rust 1.97.1 gates, 145 Python contracts,
+compileall, CodeGraph and diff checks pass. Exact pinned-nightly production coverage is
+1309/13749/17512/1452, each 100%; SHA-256
+`596097f5545490f6a4bd29a7b718a61aeb43f0ef738e02a61f372232ecd64543`.
+CI `34037184023` and MV3 `34037183844` remain queued. Actual Edge inspection of the
+published body and Checks page confirms the exact head, Draft state and queued jobs,
+with readable wrapping and no observed clipping or overlap. Writer `5559594727` is
+released. This is not product-browser acceptance or protected delivery. #267 already
+owns typed text transport and needs separate current-parent adoption and dispatch review;
+#266 itself still provides construction, not text dispatch or observed action success.
+
+The later complete queue read confirmed the canonical inventory above and all Ready
+roots remained BLOCKED, but found two new #238 review threads in addition to #147: `PRRT_kwDOTulPlM6fsIep` and
+`PRRT_kwDOTulPlM6fsIet`. Both findings were verified against `ca8f8475`: the supplement
+presented predecessor source claims as current, and the inventory assertion selected
+historical counts by formatting. Regression `6fe2af69` observed both failures. The
+repair explicitly labels the supplement historical and bounds current inventory checks
+to this latest cut, paired with the dedicated current CHANGELOG entry. A mutation test
+proves that a changed latest count cannot be hidden by unchanged historical counts.
+Thread resolution requires published repair evidence; this documentation revision still
+needs its own post-publication visual inspection and exact-head hosted checks.
+
+#### Current executable queue and lineage: 14:18 UTC
+
+Historical Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
+
+The fresh complete inventory confirms the canonical counts above. These are Ready
+review candidates, not merge-ready approvals: all remain BLOCKED. Re-fetch each
+head, base, checks, findings and counted approval before acting; do not substitute
+the historical root list below. #238's moving head remains live metadata rather than
+a self-referential SHA in this document.
+
+Current foundation #195 `63997bcf555e2c5c8e91ba287734ffba3837a1b7` remains Draft on
+`6922dd98779e8f8aad132a3b1f563d7ba6e6d070`. The earlier `48eb2d` observations are
+historical, not current lineage. #279's workflow-generation prerequisite and #212's
+sandbox-helper boundary still require their own evidence and must not be bypassed.
+
+Published transport #267 `4435ce5f561ca069c1844a1a5bd9b603505e25f7` adopts #266
+`eb6c236ff2f4a58b807a2f2c914bd1ddb6079fb3` through ordinary merge `d903cf6b`, preserving
+both histories. Actual REDs `2b960002`, `e2b49e68` and `10131eb7` preceded repairs for
+invalid deadlines, no-write preflight retention and foreign-session dispatch. The
+implementation reuses canonical timeout/session checks and sealed typed dispatch;
+ambiguous writes remain pending. Ten focused socket tests, full Rust 1.97.1 gates,
+145 Python contracts and exact production coverage 1317/13832/17596/1456 each at 100%
+pass. Coverage SHA-256 is
+`065e535fd49c34699b35d53b4cc6b09f22b3dff506f07eb5549c21b73074bdc8`.
+Actual Edge inspection verified its published body and Checks page at that head,
+with readable wrapping and no observed clipping or overlap. CI `34038399969` remains
+queued; this is not hosted acceptance, counted approval or product-browser evidence.
+Source writer `5559716889` is released.
+
+Execute the verified #238 root/lineage review repairs first, then adopt #267 into
+existing response owner #268 `8d4027e40b790d28d866051ba741db12927ec22c`. That consumer
+still uses removed generic correlation; migrate to the parent's distinct text family
+without restoring generic routing, weakening malformed-response retention, or treating
+response admission as observed action success. Received-message provenance remains a
+separate unproven boundary. Keep source and documentation writers separate. This
+documentation revision needs its own publication, visual inspection and hosted checks.
+
+#### Prior observation: 12:28 UTC
+
+All current-state wording in this prior observation is scoped to its recorded time.
+
+Observed through (UTC): `2026-09-06T12:28:16Z`. This cut supersedes volatile claims in the prior observation section below; those earlier exact-head measurements remain historical rather than transferable acceptance evidence. The complete queue read finished at `12:28:16 UTC`, after #265 publication and writer release. The new parent #264 regression is independently owned and is not yet adopted by #265.
+
+The inventory remains **125 open pull requests: 12 Ready/non-draft and 113 Draft; 13 open non-PR issues**. Protected main remains `87c4daa1830bac5a5228b6036752ad5633232085`; release and tag inventories are both empty. Ready-root source heads other than this document's moving branch are unchanged. GraphQL access recovered after the earlier rate-limit failure. GraphQL thread resolution was refreshed for all 125 open PRs alongside exact heads, bases, check rollups and formal reviews in five pages, finishing at `12:28:16 UTC`; all nested histories fit their 100-item pages, with no remaining pagination flags. #147 retains the sole unresolved thread `PRRT_kwDOTulPlM6coZwc`. #166/#220 retain `CHANGES_REQUESTED`, and the other Ready roots retain `REVIEW_REQUIRED`; all twelve are `BLOCKED`. The earlier successful GraphQL cut at `10:00:47 UTC`, subsequent failed inventory/minimal probe and complete 375-read REST fallback at `10:12:37 UTC` remain historical diagnostics, not the current freshness boundary. Active ruleset `18156473` still requires a counted approval and seven central workflows, while the collaborator inventory still contains only the author. #219 has a same-head formal bot approval, but its freshly verified GitHub decision remains `REVIEW_REQUIRED`; an uncounted review does not satisfy the gate. None of this authorizes a bypass or substitutes for an eligible counted approval.
+
+The restored foundation changes the next executable queue. #195 is Draft at `63997bcf555e2c5c8e91ba287734ffba3837a1b7`, on `6922dd98779e8f8aad132a3b1f563d7ba6e6d070`. Its owner restored product/evidence assets lost by historical whole-tree repair `5c111d0db6c363f9d1786c21cc01c5c7398007bd`; this is a content-recovery boundary, not permission to copy an old tree over later work. Fresh hosted CI `34013251657` fails in repository contracts while exact production coverage and MV3 `34013251651` succeed. The exercised failures compare inherited workflow concurrency and `nightly-2026-08-01` against the protected repository-scoped identity and `nightly-2026-08-18`. #279 owns current-workflow reconstruction; weakening the restored tests does not repair the generation mismatch. #242 remains at `2d0e9f69df9ade21d8e8e3d807c3ff644d83b310`, with a stale pre-recovery #195 base `48eb2d23009c1c804520dd5efcd0d4d072aacef1`. Its old CI success is not recovered-foundation acceptance. After the owner prerequisite, adopt the verified foundation content-aware and non-destructively before regenerating descendant checks.
+
+The owner has already moved #255 directly onto #252 so the valid #253/#254 deltas and connection-bound repair travel together. #253/#254 remain open, not silently discarded. Current native Rust and coverage checks succeed on each of these exact heads; earlier local coverage counts do not validate these newer heads:
+
+| PR | Current head | Exact native CI |
+| --- | --- | --- |
+| #255 | `6865faa8185c5fe6a8b6aaba9535d774575b0061` | `34003175242` success |
+| #256 | `881c7f09ee9161ce8664dd75226938ecf60b85e5` | `34003259914` success |
+| #257 | `8f1507346f65798a6bf4eaf370d65a2d406a6f44` | `34003269343` success |
+| #258 | `b8eaa97f7a417c79250b6b760ff214ac03d39b8a` | `34003285350` success |
+| #259 | `91d95423cf31947f691db5ebbd3072c481d86542` | `34003299669` success |
+| #260 | `e5228396be8d9faade44a30aed704cacbeb91b46` | `34003322381` success |
+| #261 | `572fc4224ddc09c010bd9ccf100764076899495a` | `34003334868` success |
+| #277 | `973e34bc24ae9bdd96a50764f2be6c8603eff66c` | `34003345253` success |
+| #263 | `3f22de94b63da83eaa8b5b1270912b21a3ecd006` | `34009256997` success |
+
+#263 now targets actual #277. Its current Rust and exact-coverage jobs `101422055630` / `101422055538` succeed after the owner repaired formatting and exercised the unsubscribe frame-error branch. This is native exact-tree evidence, not protected integration, authenticated browser post-condition, or central review acceptance. The deeper #195 foundation prerequisite still applies to this whole stack.
+
+#288 is now Draft at `fd589cd693946ef1ce2c9270c2dfb6a1087bdfb9`. The independently inspected two-file diff from `0f434bc...` contains documentation-consistency RED `e74acd7...` and the doctoring correction `fd589cd...`, with no production or workflow change. Its sandbox-source and historical mock-fixture boundaries are explicit. Current Strix and Noema succeed, CodeQL fails, and native Rust/coverage/MV3 are skipped; predecessor local 173-test/coverage proof is not current browser execution. The earlier planned documentation repair is done and must not be duplicated.
+
+#148 is now Draft at `0135984f1bc1f68d89d7777f49c4999474105a12`. Current native CI `33990522263` succeeds, but pinned-Chromium MV3 `33990522248` fails. The later bounded diagnostic artifact locates browser-crash failure at session creation; #212 records 0/3 for every real-browser lane after all sandbox-disabling overrides were removed. #43's separate helper-generation MV3 `33866932365` succeeds only on its own exact source. #212 owns sandbox-helper adoption against the current protected workflow, followed by fresh consumer execution; cleanup bookkeeping and leaf success are not transferable runtime evidence.
+
+#150 is Draft at `6ab7fc9166f397f49442243492ae88d6ccad56cf`, ordinarily adopting #148 `0135984f1bc1f68d89d7777f49c4999474105a12` while preserving its combined ordinary-pass teardown waiter. Reused parent regressions first exposed three unsafe browser-launch paths and missing crash-stage diagnostics on the child predecessor. The repair preserves all sandboxed parent launch/cleanup behavior and the child waiter delta. All 253 Python contracts pass on Linux without skips; the corresponding macOS run explicitly skips three Linux pidfd cases. Host/guest tracked manifests match `985411ef6a7d3b812be086b43b6341a96f6b8acace63c0161fa8453aae065bd3`. Complete local Rust gates and 415/3555/4444/476 exact 100% coverage pass. Native CI `34018588292` now succeeds, but MV3 `34018588298` / job `101446865332` fails on that same head. Artifact `9985228944`, digest `sha256:05b6d1cd6092a9e2e18a5003a5a1faeb73b8a8d9b4fd7533f7f24bc938833436`, records all four browser lanes at 0/3, with each browser-crash trial failing at `session_create` and all profiles cleaned. This identifies the failed stage, not the underlying Chrome startup cause. The bounded diagnostic and sandboxed runner/helper follow-up is attached to [#212 comment 5557994915](https://github.com/ContextualWisdomLab/OriginWeave/issues/212#issuecomment-5557994915); no sandbox bypass or unchanged rerun was attempted. #147's ordinary-deadline review thread remains unresolved until this child is actually integrated.
+
+#### Subscription provenance predecessors
+
+The following paragraphs preserve exact predecessor observations. Their remaining-boundary statements and queued checks describe those revisions, not the newer original-registry repair below.
+
+#264 remains Draft on actual #263 `3f22de94b63da83eaa8b5b1270912b21a3ecd006`. The preceding published head `43d3b5a3a2b5ce4f51a93d1152a0ee82620f4f3e` retained private original-command identity after test-first `73f11de2232060ac7680e188db88ef7609123296` and the two-registry collision regression. Its five admission tests, 144 Python contracts and 1273/13294/16963/1430 exact 100% local coverage are predecessor evidence; native CI `34020055803` is now cancelled, not current acceptance.
+
+The inbound-provenance repair was first preserved as unpublished checkpoint `bd29f405a6c927b2f7d1c437dccbfb8bcec424f1`, following real crossed-connection RED `918c4ebe`. It reused the existing connection-owned reader and private connection generation to reject foreign success, protocol-error and navigation-event messages before pending-command or document mutation, then proved original-connection recovery. Seven admission tests, 23 focused tests, all 144 Python contracts, full Rust gates and 1273/13311/16973/1432 exact 100% coverage passed on that private tree. Independent read-only review found no actionable defect; it is not counted approval. Those measurements remain attached to that predecessor, not transferred to the later integration below.
+
+A separately active writer advanced the remote through `bc2e69d5`, `a3dfd190`, `c46642e2`, `ed434fc7`, `d8983d50`, `ced211a7`, `7ae9657c`, `7c20907b` and then-current head `8ebcc6131a5dd6bf0b4720c2f1ff8d40d1cc39f8`. The inspected intervening commits contain valid regression and connection-provenance work. [Writer coordination comment 5557964460](https://github.com/ContextualWisdomLab/OriginWeave/pull/264#issuecomment-5557964460) requested task identification, writer acknowledgement and a final exact-head handoff. The private writer claim was released without a competing push; the subsequent integration and release are recorded below. The current caller-supplied registry binding, actual-resend freshness on the same connection, unsubscribe transport/lifetime provenance and real-browser causality remain distinct unproven boundaries. The #195/#279 prerequisite, branch-instrumentation warning and Draft status remain; no workflow, dependency, deadline or gate changed.
+
+A fresh unmodified detached checkout at the earlier remote `8ebcc613` diagnostic cut passed Rust 1.97.1 network/all-targets checking, strict workspace/all-targets/all-features Clippy and all 145 Python contracts. At that cut, the Rust 1.97.1 formatting check fails in five test/support files. The pinned nightly coverage test run completed, but the unchanged numerical verifier failed at lines 13327/13330 and regions 17004/17008; functions 1274/1274 and branches 1434/1434 are exactly covered. Coverage artifact SHA-256 is `f352e3f04bdc3ed1912be7c452c5c426d9caaec7dd1a613b7d3ab9c6486ab69e`. The uncovered regions were the duplicated missing-generation propagation in correlation completion at line 240 and the admission diagnostic at lines 396–397. Independent source review confirmed that the former follows an intent check whose retained constructor always sets the generation; shared validation and a real rejected-event diagnostic/recovery test addressed these causes without exclusions or fabricated production state. The comparison found no reachable authority bypass and identified remote typed-send malformed/unknown/error fixtures, stronger rustdoc and the source contract as deltas to preserve together with private original-connection recovery, unchanged document state, authentic receipt redaction assertions and bounded evidence. Neither private nor remote results establish combined-source acceptance. Earlier native CI `34021674375` was queued at that diagnostic cut; formatting, numerical coverage and hosted execution remain separate gates. The coordination comment recorded the diagnostic and preservation map before shared-source integration or writer acknowledgement occurred.
+
+After the other writer's explicit [release at 52cff969, comment 5558261278](https://github.com/ContextualWisdomLab/OriginWeave/pull/264#issuecomment-5558261278), ordinary content-aware merges preserved both complete contributor histories. Published #264 head `2a9fdc5418b9af10353cdad0f6f6470655bf457d` contains private `bd29f405a6c927b2f7d1c437dccbfb8bcec424f1` and remote `52cff96954c3fec7b8cda5409e4560e970bfcbdf`: both complete lineages are now ancestors. Independent read-only review found no actionable preservation defect. Final-tree Rust 1.97.1 formatting, locked workspace/all-targets/all-features check/test/strict Clippy, strict rustdoc and all 145 Python contracts pass; the workspace run retains all 26 focused socket tests. The unchanged pinned-nightly verifier confirms 1273/13317/16984/1432 functions/lines/regions/branches at exactly 100%, with coverage artifact SHA-256 `f7e71e6fd67723688535053c389b4ff718c4563b403924af462848d3d51b541f`. No test exclusion or quality-gate change was introduced. The normal push and exact remote head were verified, and the integration writer released. Fresh exact-head native CI `34024499232` remains queued; this is local combined-source acceptance, not hosted acceptance, counted approval, protected-main delivery, browser causality or release evidence. The unstable branch-instrumentation warning and all unproven boundaries above remain.
+
+#### Original-registry ownership predecessor
+
+The preceding published #264 head is `10f138f8787d596e8b556fe50c9e4e52bc1295b7`, still Draft on the same #263 base at that cut. Four real-socket REDs at `b3ffeac9` proved original-command send, original-receipt admission, original-event admission and document mutation could use a replacement registry with colliding local identifiers. The document-mutation case did not require matching external context text. The repair retains one opaque core-owned allocation witness through the existing command, binding and subscribed observation, rejecting foreign ownership before correlation/I/O, replay insertion or the common document-mutation sink. Origin binding uses that same sink. Original registry moves and cloned witnesses remain valid; owner destruction does not let a replacement inherit the old identity. The socket receipt regression drops the original registry before creating its replacement, and the existing core diagnostic contract exercises the real dropped-owner mismatch.
+
+The final exact head passes all four new socket regressions, all 145 Python contracts without skips, complete Rust 1.97.1 formatting/check/tests/strict Clippy/rustdoc, compileall, graph and diff checks. Unchanged pinned-nightly coverage is exactly 100% at **1278/13371/17056/1434** functions/lines/regions/branches, artifact SHA-256 `9f5b4942d6a040a83f09593a2b81406c44bc9d2b402fa995a95294ce8e9cbb7e`. Initial `e686b3a0` coverage missed one diagnostic line and three regions in the core unit copy; the genuine error path now exercises that copy. Intermediate `33787617` reached 100% but failed the strict test-lint rule; the final test uses the established error-collection/cardinality assertion style. No exclusion, lint allowance, dependency or production authority relaxation was introduced. Independent read-only review found no actionable authority, lifetime, construction, preservation or documentation finding. Both complete contributor lineages remain ancestors; normal publication, the current PR body and explicit writer release were verified.
+
+Native CI `34026519860`, Rust `101468348712` and coverage `101468348775`, and real MV3 `34026519878` / job `101468347302` now all completed successfully on `10f138f8`. These are terminal-success predecessor evidence, correcting the earlier queued observation without validating a later head. The repair pins the original registry but does not authenticate the registry-to-transport association. Actual-resend freshness was still unfinished on this predecessor; the next section records its separate repair. Unsubscribe lifetime/transport provenance, action causality and end-to-end Chromium product acceptance remain unfinished. ADR 0107 stays Proposed, #195/#279 remain prerequisites, and the unstable branch-instrumentation warning remains. No workflow, gate, merge, tag or release mutation occurred.
+
+#### Command-response freshness predecessor
+
+The preceding published #264 head `805051527cf95e14ba126c9dd3159db86d190224` was Draft on unchanged #263 `3f22de94b63da83eaa8b5b1270912b21a3ecd006`. Actual success-first RED `92fd0b07` and error-first RED `15aea15e` each exposed four repeated-send lifecycles: completed, retired, replacement correlation and unread buffered response. Three additional actual-wire REDs exposed raw/typed mixing and repeated identifiers; a compile-fail contract exposed the pre-upgrade socket alias. The shared established owner enforces strictly increasing typed identifiers and exclusive raw-text/typed mode across all five typed senders. Zero is valid first, out-of-order replies for distinct outstanding commands remain valid, and reader reconstruction/Pong cannot reset history. The consuming socket handoff remains; its nonconsuming alias is removed. This is a stricter local dispatch policy, not a W3C requirement or authenticated browser-session association. The next section records its later teardown refinement; this predecessor's unfinished boundaries and screenshots are not current-head acceptance.
+
+All 22 focused admission/frame tests and the added real public opening-deadline test pass within the complete Rust 1.97.1 workspace verification. Formatting, locked all-target/all-feature check, strict Clippy, warning-denying rustdoc, all workspace tests, all 145 Python contracts, compileall, CodeGraph and diff checks pass. Unchanged pinned-nightly coverage is exactly **1279/13409/17087/1436** functions/lines/regions/branches, artifact SHA-256 `64a62b1e03ee3ed3d62654a3227495e82b7ff93357038ae146d8584c581ac060`. Earlier buffered-fixture BrokenPipe/reset failures and `e6c02cf`'s 17086/17087 region result remain recorded. The latter combined complementary ordinary-library/unit-test gaps rather than unioning source regions; the real public deadline case covers the ordinary copy, retaining the private real revoked-socket test without production or verifier changes. That one-nanosecond case has local passing evidence; portability beyond executed targets remains unproven. Independent read-only review found no actionable finding and does not count as GitHub approval.
+
+Fresh exact-head CI `34029687813`, Rust `101476824185` and coverage `101476824305`, and MV3 `34029687816` / `101476824298` are queued. Post-publication actual Edge visual inspection covered the new PR body and Checks page: exact head/base, Draft state and three queued checks matched the API; desktop screenshots showed readable wrapping with no observed clipping or overlap. GitHub presentation evidence is not OriginWeave product-browser acceptance. Both contributor histories remain ancestors, ordinary publication was verified, and the #264 writer was explicitly released before this documentation-only slice. Registry-to-transport authentication, unsubscribe lifetime/transport provenance, action causality, real browser acceptance, #195/#279 integration, protected delivery and release gates remain open.
+
+#### Subscription teardown ownership and provenance predecessor
+
+At the preceding cut, published #264 was `7fb93e77b800f27187a5c02333298cc31a025bd6`, Draft on unchanged #263
+`3f22de94b63da83eaa8b5b1270912b21a3ecd006`. Lifetime RED `2c45cea8` reproduced event admission
+after borrowed teardown construction. Transport RED `ce6f6fd4` reproduced 91 actual masked bytes
+on a foreign connection and foreign success/error consuming the original pending command. Genuine
+original replies then failed as no longer outstanding; real servers were joined before assertions.
+
+The repair consumes the existing non-cloneable subscription receipt and retains its original
+connection identity through teardown dispatch and acknowledgment. It reuses connection-aware
+correlation and the existing connection-bound reader; no revocation registry, dependency, workflow
+or quality-gate change is added. Receipt reuse is now rejected by the compiler with `E0382`, not
+merely by a fixture setup error. Construction failure also ends local admission availability.
+Previously admitted observations are not retroactively revoked, and acknowledgment does not prove
+event drainage or browser cleanup. Proposed ADR 0107 records these costs and rejected alternatives.
+
+All three transport regressions, eight preserved failure cases, fourteen admission tests and the
+escaped-identifier round trip pass. Full Rust 1.97.1 workspace/all-feature formatting, locked check,
+strict Clippy, warning-denying rustdoc and tests pass, including three network doctests. All 145
+Python contracts, compileall, CodeGraph and diff checks pass. Unchanged pinned coverage is exactly
+**1279/13421/17106/1438** functions/lines/regions/branches; artifact SHA-256
+`277c5bc0a17f16966a8eb388d54e4e172be36356d7dead5bbead9ed53ad8cff6`. The unstable branch warning
+remains. Independent read-only source review found no actionable issue, not counted GitHub approval.
+
+New exact-head CI `34031281825`, Rust `101481131938` and coverage `101481131907`, and MV3
+`34031281812` / `101481131930` are queued, not GREEN. Both contributor histories are preserved,
+normal publication was verified, and source writer `5557964460` was explicitly released before
+this documentation writer began. The required new-head visual inspection remains incomplete:
+actual browser control reports that the Mac is locked and automatic unlock is unavailable. Manual
+unlock was requested; neither the previous #264 screenshots nor the previous #238 Preview validates
+these new heads. Registry-to-browser authentication, action causality, real Chromium acceptance,
+#195/#279 integration, protected delivery and release gates remain open.
+
+At this read, #238 predecessor `81e334aff063290bde3298671cfd20e77c7c1fff` still has queued native
+CI `34030165220` plus central review/security work. This documentation commit needs its own checks;
+neither its earlier 172 local Python passes nor #264's source coverage transfers to the new head.
+
+#### Current pointer integration and independently owned parent repair
+
+Published #265 is `35555d0f8491d4ee95c2e61d1a2aaa2c02a0635c`, still Draft. Ordinary two-parent
+integration `7535d8af` preserves original child `ffa70ee0f499b86ff51837fb95733fd5cf57ff89` and
+parent `7fb93e77b800f27187a5c02333298cc31a025bd6`. Real-socket RED `3f9cdc1a` observed no command
+bytes but one outstanding command after a zero deadline. The repair retains the child's immediate
+admitted-node revalidation and the parent's typed dispatch, local no-write retirement and ambiguous
+write retention. Both use the existing registry identity allocation. Independent review identified
+four silently deleted parent postcondition tests and two obsolete sender calls; all were restored
+or migrated without dropping assertions. The final review found no remaining actionable finding,
+which is not a counted GitHub approval.
+
+All 19 focused tests, complete Rust 1.97.1 formatting/check/tests/strict Clippy/rustdoc, all 145
+Python contracts, compileall, CodeGraph and diff checks pass. The exact-head pinned-nightly rerun
+passes unchanged production coverage at **1293/13567/17273/1440** functions/lines/regions/branches,
+each 100%; artifact SHA-256 is
+`b1374f65538fb6557b4e5d877aa067e183202ed1f56c75be047b3ff2dfdc92a0`. This is local source evidence.
+CI `34032661775`, Rust `101484986934` and coverage `101484987071`, and MV3 `34032661781` /
+`101484986948` remain queued. Normal publication was verified; writer `5559058158` released.
+
+The live parent #264 subsequently advanced to test-only `b4702cd503fa3f721e0d1f44b355563753dac0a2`.
+Its 129-line socket regression targets registry-session A dispatch over transport-session B;
+writer [5559076006](https://github.com/ContextualWisdomLab/OriginWeave/pull/264#issuecomment-5559076006)
+released in [5559140277](https://github.com/ContextualWisdomLab/OriginWeave/pull/264#issuecomment-5559140277)
+without executing the regression or implementing a production repair. Parent CI `34031977586`
+and MV3 `34031977597` remain queued. The available local backend can now execute that regression
+under a new source lease. Adoption waits for verified repair: #265's local GREEN is **not latest-parent acceptance**.
+The PR base snapshot still names `7fb93e77`, so it must not replace the fetched live parent as
+dependency evidence. No parent source was changed during this documentation slice.
+
+Browser access recovered after the earlier locked-screen failures. Fresh actual Edge visual
+inspection of #265's published body and Checks page verified head `35555d0f`, Draft state and the
+three queued checks. The inspected desktop screenshots show readable wrapping with no observed
+clipping or overlap. This replaces the earlier lock blocker only for that inspected presentation;
+the new documentation head still requires its own visual inspection. GitHub presentation is not
+OriginWeave product-browser acceptance, browser-process authentication or navigation causality.
+
+The #238 predecessor `88447b7d43e4a87b48fb8fde33f93b2e3ce8cbc5` still has pending native CI
+`34031539064` and central review/security work. Its 172 local Python passes and #265's source
+coverage do not transfer to this documentation commit. The next executable items are this dated
+evidence update, then an executed parent regression and verified repair under a new writer lease; #195/#279,
+#212 and protected-main/release gates remain prerequisite work.
+
+#### Scheduling and central-owner follow-up
+
+This section retains owner-reported scheduling and central-repository evidence. Those reports are
+not refreshed owner-source or released-consumer acceptance in this bounded documentation slice.
+
+The existing hourly-task owner attempted a prompt-only coordination/evidence improvement, but the scheduler rejected the update with `too_many_active_automations` at the ten-active-task limit. The owner reports that a subsequent read confirmed the old prompt and update timestamp, cadence, enabled state and notification preferences were unchanged. This is a failed update, not an applied scheduling improvement; no duplicate schedule or unrelated-task pause was created.
+
+On #238's earlier `a2600ee8...` head, native CI `34018002926` succeeded. The later `38e5eb44...` cut recorded queued CI `34021771606`; subsequent published `7ca86a2f546adf3a0af3fec41d97675115db384d` still had queued native CI `34022635038` before this refresh, together with central review/security work. At the now-predecessor `8e6284f5f89965983d025ce302d7c99e2f5c38d5`, native CI `34027019757`, Rust `101469694749` and coverage `101469694911` have completed successfully; nine central security/review jobs remain queued. This document's next commit requires new exact-head checks; it cannot inherit those results. Existing central CodeQL dispatch/verdict work remains with `.github#712`. Central Strix #1563 remains unmerged at `eaf9594f7fe8d8e1994349289183d6cbad056579`, reported behind its base: the earlier `13fbb48e...` source repair and owner-reported harness results do not prove consumer recovery. The unrelated #37 local 900-second deadline remains distinct from the #166 completed-report classifier defect and #219/#240 actual gateway/hosted-limit incidents. The active central Noema owner additionally reports source REDs under `.github#1641`; owner reports and comments are not released transport or SDK-compatibility evidence, and no consumer workaround is introduced here.
+
+### Prior observation cut: 2026-09-05
+
+The following cut preserves prior source measurements and diagnostics. Present-tense wording within this dated cut describes its observation time only; use the latest verified cut above and live APIs for the current queue.
+
+Observed at (UTC): `2026-09-05T13:50:14Z` (full REST inventory/review/check refresh and targeted source verification; subsequent bounded diagnostics are recorded below).
+
+GraphQL rejected both the inventory query and a minimal probe with its rate-limit error while the REST endpoints remained available. REST reviews, check runs, and combined statuses were read for all 125 open PRs; truncated output was recovered by querying the exact missing records. New source heads were refreshed separately after pushes. Review-thread resolution cannot be inferred from these REST endpoints: earlier thread-resolution statements below retain their prior observation date and are not refreshed merge evidence. No query failure, queued check, or metadata update relaxes a quality or review gate.
+
+- #37 Strix run `33946242342` / job `101285198427` failed at 13:26 UTC. Artifact `9970303662` (SHA256 `93ce81a9c3060f52a23d0910b5a995a2b491aeb3889a54432d954747a97010ca`) records `Strix run timed out after 900s`, 903 seconds elapsed and exit 124; its current-attempt record is interrupted, not completed. The provider-unavailable wrapper label does not establish a provider outage. Canonical Strix deadline/classification ownership received the exact artifact; no consumer timeout change or blind rerun was made.
+- #148 hosted compatibility run `33962062608` / job `101295542349` failed at 13:33 UTC on `bded4fc9d32e5e047cea99d182aa05ae2cd03bf6`: browser-crash recovery fails 0/3 trials, while ordinary Agent Task and forced-close each pass 3/3 and profiles are cleaned. Artifact `9970408851` (SHA256 `daefc93ff2792cbc82a8937107e7992e91447f56692c4ea7071bc2b39c0faa7b`) initially exposed only `RuntimeError`; the stage was not reconstructible from it. Subsequent owner diagnostics and newer all-sandboxed results are recorded in the latest cut, not backdated into this earlier artifact.
+
+- Protected `main` is `87c4daa1830bac5a5228b6036752ad5633232085` through #286. PR #286 skips repository-native CI jobs for draft pull requests; a skipped job is not passing evidence.
+- The September 5 search also returned 125 PRs (12 Ready, 113 Draft) and 13 non-PR issues. PR #290 was converted back to Draft after current-protected workflow review found that its #245-relative patch would drop protected #286 lifecycle controls; queue movement is not protected-main delivery. PR #248 also returned to Draft for current-parent adoption and complete revalidation. Temporary synthesis #291 has since been externally merged and closed as recorded below; its closure does not establish product acceptance.
+- Ready roots are #37 `1e2f41072854edcdbaf0f9ecf14697a3bfd62195`, #50 `ad87cfea59db711cb29ef90559790ba77e22029f`, #166 `e84a1a2cc82b1c666218efd441da97849f47b8c2`, #219 `65e4315d80137badc0b55e1b9617015beb1db568`, #220 `e545b94e1de499b96b867694f80ac04ad247becd`, #229 `024f63690cf05cfe6f0d4a430f0e18ea8fd2c4d6`, #240 `24930a3a9ee79c0b712ee3df6589b0592eb6e18f`, #272 `b1cae8ad1cbd8eb6992037c830aea30b9aa436b3`, #274 `802d0bdff7536d9ac253305d3e0237b4e4a1789e`, #285 `f455c2cd64b3dd3f027c91d396103792a205ddd0`, and #287 `af83c40dd2990a03064a92ca75430a9cc400f098`. PR #290 is Draft at exact head `ebeefcd534db4324498fdb18046ebc6255ddcdf2`, remains a #245-dependent workflow-owner candidate, and is not a root. PR #238's moving exact head is intentionally omitted from its self-referential document; live PR metadata is authoritative. Their hosted exact-head checks remain non-terminal and none has an eligible exact-head approval, so none is merge-ready.
+- PR #166 exact head `e84a1a2cc82b1c666218efd441da97849f47b8c2` and PR #220 exact head `e545b94e1de499b96b867694f80ac04ad247becd` retain formal `CHANGES_REQUESTED` decisions. REST review records were refreshed; thread-resolution state could not be refreshed because GraphQL rejected the current read. The last successful thread-resolution snapshot was the prior 12:36 UTC sweep, not this REST refresh. Thread resolution alone is not approval or grounds to dismiss a review, and the active counted-approval gate remains unmet.
+- Targeted CI RCA at `2026-09-05T06:45Z` confirmed that #219 Noema run `33925442322` / job `101226941175` and #240 run `33925596923` / job `101227537529` had terminated on gateway HTTP 502 without review verdicts. Neither unchanged exact head had a successor Noema run. One supported failed-job rerun per head created attempt 2, with queued jobs `101264704581` and `101264705575`; Noema attempt 2 is queue-admission evidence, not provider recovery or a review verdict. Successful sampled Noema wrapper runs skipped their review jobs and cannot establish recovery. The CodeQL failures independently record dispatch handoff: #219 central runs `33947047322`, `33947036734`, `33947037072` and #240 runs `33947189162`, `33947189634`, `33947187643` match those exact heads; these central CodeQL dispatches remain queued and were not duplicated. Active Strix runs were left untouched. Revisit these exact attempts after a terminal result; do not create repeated retries while they remain queued.
+- A fresh repository Actions query at `2026-09-05T06:12:27Z` returned **115 queued workflow runs**; the newest sampled run was #238 CI `33949183271` on predecessor head `bb2f18570f319a763acc24d7c4206ec9d0a11a29`. This is runner-admission/backlog evidence, not a code failure or passing check; rerunning unchanged heads would only add duplicate queue load. The moving PR's exact current head remains live-metadata-only to avoid a non-convergent self-reference.
+- Active organization ruleset `18156473` (`CWL Central required workflows`) applies to the default branch. It requires one approving review, dismissal of stale approvals after pushes, resolved review threads, extra approval for unattributed changes, and **7 central required workflows**: `opencode-review`, `pr-review-merge-scheduler`, `security-scan`, `strix`, `sast-semgrep`, `noema-review`, and `codeql-pr`. Current `ContextualWisdomLab/.github` consolidates OSV and Scorecard PR scanning into the required `security-scan.yml`; standalone `osv-scanner-pr.yml` and `scorecard-pr.yml` are no longer ruleset entries. Administrative bypass capability is not authorization to use it.
+- PR #284 head `61bcf88c960c6c437ccd29b3fbb73cd4325f9e5a` reached `main` through rule-suite `3948421709`, whose live result is **`result: bypass`** with actor `seonghobae`. The required non-author approval and workflows were not proven before integration. The earlier pending post-merge state is historical: fresh native Rust and production-coverage checks now succeed on `87c4daa1830bac5a5228b6036752ad5633232085`. This is a governance incident tracked by #215, not a policy-compliant merge or evidence that later checks can retroactively authorize it.
+- PR #285 is Ready at exact head `f455c2cd64b3dd3f027c91d396103792a205ddd0` on current protected main. It addresses #284's three post-merge review findings without changing `.github/**` and adds a dedicated regression contract. All 154 repository contracts, the full Rust gates, and exact 100% local coverage pass. Its Ready transition materialized repository-native CI `33930234387`: Rust contracts `101207153048` and Production coverage `101207153244` succeeded by `2026-09-05T03:16:29Z`; native CI success does not replace the seven required central workflows. The Ready event did not materialize fresh central required-workflow runs. Targeted RCA at `2026-09-05T06:53Z` verified that the previously cancelled Security Scan `33924016851`, SAST Semgrep `33924016903`, and CodeQL PR `33924016883` still bind the current head and protected base, with no duplicate central CodeQL dispatch. The convenience CLI failed on a cross-repository workflow lookup before rerunning anything; the supported run-ID REST route then created attempt 2 for all three, with queued scope/language jobs `101265706025`, `101265708882`, and `101265710202`. These are verified replay admissions, not terminal scan results or proof that Ready-event materialization is repaired. Remaining required review evidence and the absent eligible approval still block merge; toggling Draft or creating a no-op commit is not an acceptable substitute.
+- Issue #279 owns the remaining documentation-CI partitioning gap. PR #287 is Ready at exact head `af83c40dd2990a03064a92ca75430a9cc400f098` on current protected main; it is the workflow-independent classifier foundation, changes no `.github/**` path, and is the prerequisite for an authorized workflow owner. Its predecessor failed its own release-record contract; the current head binds Git rename/copy similarity to blob identity, and its focused and full local suites plus exact 100% Rust coverage pass. Exact native CI, Semgrep and scoped Security Scan have now completed as described next; remaining required review/CodeQL evidence and eligible approval are not complete. PR #282 is Draft at exact head `b64e0708584beff3fb54acf226cb3e667773e473` on current protected main. Its latest workflow/test delta removes `converted_to_draft` and `closed` from the CI event list, removes #286's closed-event guard, and makes the repository contract assert those protections are absent. Exact review `5120043505` therefore requires the authorized #279 owner to reconstruct the valid trusted-base classifier/contract partition while preserving #286's lifecycle and closed+Draft fail-closed controls; Draft checks do not transfer as GREEN.
+- #287 CodeQL handoff RCA at `2026-09-05T08:34Z` confirms that exact native CI `33931806137` passed, Semgrep `33931806165` executed its scan successfully, and Security Scan `33931806226` executed Scorecard/Trivy while scope-skipping gitleaks/OSV/dependency-review. CodeQL wrapper `33931806139` failed intentionally after its three language jobs recorded successful dispatch and a pending verdict, not a failed terminal scan. The cross-repository convenience CLI returned 404; direct REST job logs identify the handoff. Central `.github` dispatches `33954721186`, `33955024697` and `33955164029` have display titles identifying the exact OriginWeave target head, but their immutable run heads are central-owner revisions (`71dd84d...` / `27d7331...`), not the product source head. All three remain queued and were not duplicated; acceptance still requires validated target checkout, an authenticated terminal verdict and the original-job replay. OpenCode is queued, Noema/Strix remain in progress, and no eligible formal approval exists.
+- PR #290 is Draft at exact head `ebeefcd534db4324498fdb18046ebc6255ddcdf2` on #245 exact `a769f484e2c110e0523b3b28cd21573f43867562`. Its workflow/repository/PR-number concurrency isolation and manual-dispatch non-cancellation intent are valid, but its predecessor-relative MV3 patch would remove protected #286 `converted_to_draft`/`closed` lifecycle events and the closed-event job guard. Exact review `5120039692` and canonical-owner issue #212 comment `5549868655` require reconstruction on the current protected MV3 workflow; the scheduled product writer did not mutate `.github/**`.
+- PR #283 is Draft at exact head `c904300a6a1bda83af24f84d586f1c5f6a6491aa`, stacked on predecessor #282 head `b54a5856d8201911f05d69622f0d5594a371adf0` rather than current #282 exact `b64e0708584beff3fb54acf226cb3e667773e473`. It must adopt the corrected current parent before its compare can become current evidence. The child remains one prose-only doctoring canary; Draft-hosted jobs are skipped, so no trigger-shape GREEN is claimed for #279. Runner admission is a separate condition.
+- PR #288 is Draft at exact head `0f434bc29d468127412366b6864b733b40b83c4d` on current protected main. Predecessor `39e36256651f62940ec3ca6149067f0cfcb2285a` had 172 passing Python contracts; that evidence is historical. The subsequent test-first sandbox contract and doctoring change led to the ordinary launch override removal, followed by restoration of the original bounded ChromeDriver diagnostic. Direct cumulative comparison from `99fea8985b4649c444ecc72c11e72c0693b17f5b` to current `0f434bc...` confirms one net production line removed: `--no-sandbox`. Fresh isolated verification confirms all 173 Python contracts pass on that unchanged head, including the source contract for both ordinary and Agent Task launch paths. Complete Rust gates and numeric 100% coverage also pass: 521 functions, 4420 lines, 5349 regions, 654 branches; the unstable branch-option warning remains. The doctoring problem paragraph still describes the historical override as current, so its existing author owns a bounded documentation/test correction after this writer releases. Unit-fixture output is not executed browser evidence. #212 still owns authorized workflow activation and fresh sandboxed pinned-Chromium compatibility/Agent Task trials; no workflow, sandbox-helper configuration, or hosted replay was changed here.
+- Issue #28 remains the P0 governed-browser integration target. PR #261 is Draft at exact head `934eb7d37568b439c442ffe1d1f6a9c8f8ed58a0`; #261 now includes actual #260 `2c5049aff97a90958e8262b1d403bdcbd64a1e8b` by normal merge. Its origin-binding implementation and tests remain byte-identical to predecessor `323ac9e147691e9f6572711f5a748e13f1036624`. Native release-contract discovery reproduced zero tests before parent adoption and one passing test after it. Fresh verification passed 18 focused loopback tests, 142 Python contracts, full Rust gates, and numeric 100% coverage: 1173 functions, 12104 lines, 15511 regions, 1334 branches. Exact CI `33967462582` is queued; no previous check is transferred.
+- Repair PR #277 is Draft at exact head `117f6414e8a6db46eb2b32f4ebae85cf2a208371`, based on actual #261 `934eb7d37568b439c442ffe1d1f6a9c8f8ed58a0`. Parent adoption `3c0484174eeda0703492ba76b530be125e3e99dd` preserved its two subscription production files, three test files, and Proposed ADR, with 16 focused tests, 142 Python contracts, full Rust gates, and numeric 100% coverage at 1220 functions, 12774 lines, 16397 regions, 1418 branches. The subsequent deadline repair reused the existing frame validator before registration: its RED observed no subscription bytes but two outstanding commands where only one pre-existing command should remain. Zero and over-limit deadlines now preserve that command and leave the rejected identifier reusable. Post-registration frame-failure retention remains conservative, including the separately tested repeated-mask rejection. Fresh current-tree verification passed 11 focused tests, 142 Python contracts, complete Rust gates, and numeric 100% coverage: 1221 functions, 12781 lines, 16404 regions, 1418 branches. The actual typed family is `NavigationCommittedSubscription`, not the historical proposed `SessionSubscribe` name. The response parser and authority boundaries are unchanged; exact CI `33968278171` is queued, and `warning: --branch option is unstable` remains. Stale conflicting #262 is not superseded, and #263+ are not retargeted or promoted until current-parent hosted acceptance and complete successor evidence exist.
+- Stacked PRs #178 and #85 were repaired without force pushes at exact heads `640e594dbc0a64f251a0f28b8e80943f53337e40` and `c9adea680d6186f1842a280e248ce55da4f1305b`. Parent-relative deltas now preserve the current MV3 diagnostic and extension-authority contracts; local full suites and exact 100% coverage passed, while hosted exact-head acceptance remains independently required.
+- PR #70 is Draft at exact head `77eb0f2ee71783e06171784b7173c0b4cd530e61`; it targets protected main but has not adopted current main `87c4daa1830bac5a5228b6036752ad5633232085`. Its focused Agent Task contract proves the pre-repair `--no-sandbox` launch as source RED and the branch removes that argument only from the Agent Task lane. Sandbox-enabled pinned-Chromium E2E and repository-wide GREEN remain unproven while its exact-head browser/CI/security runs are non-passing.
+- DDD/MCP repair #272 is Ready at exact head `b1cae8ad1cbd8eb6992037c830aea30b9aa436b3` on current protected main. Its exact-head hosted acceptance remains incomplete and no eligible approval exists. Documentation child PR #273 is Draft at exact head `e5c8fcb66bf644dfa750bb1b40ba3d600cb7805a` on predecessor #272 head `fe124e447cad3f679e22337fb6fbdfd135ab3652`, so it must adopt the parent only after #272's current head completes its gates. Both remain active-PR evidence.
+- PR #229 is Ready at exact head `024f63690cf05cfe6f0d4a430f0e18ea8fd2c4d6` on current protected main. Its 157 Python contracts, full Rust gates, and exact 100% local coverage pass, while hosted exact-head acceptance remains incomplete and no eligible approval exists. The presentation-identity work retains the boundary that Chromium application and page-observed effectiveness remain separate adapter/browser-E2E work.
+- PR #281 remains Draft at exact head `adaca6427d68f550b39293a69b7c733430d1c385` on canonical MV3 parent #43 `6f3134d18d3118aab33d28048671dc71a5f47b77`. Its child delta is diagnostic-evidence contract/doctoring only; inherited #43 implementation redacts browser/page-derived HTTP and WebDriver protocol data, startup exception details, returned capability values, DOM-dataset values, and click post-condition text before CI/audit evidence. Exact-head CI and Manifest V3 Compatibility succeeded and review threads are resolved, but parent-first workflow-ownership verification remains open, so the child stays Draft.
+- PR #43 is Draft at exact head `6f3134d18d3118aab33d28048671dc71a5f47b77`; it targets protected main but has not adopted current main `87c4daa1830bac5a5228b6036752ad5633232085`. The branch now installs and configures the pinned `chrome_sandbox`; exact-head hosted verification remains required, and restoring `--no-sandbox` is not acceptable.
+- PR #274 remains the bounded GitHub Pages source/README/CHANGELOG public-surface delta at exact head `802d0bdff7536d9ac253305d3e0237b4e4a1789e`, with a repository regression contract for the exact badge target, pre-alpha status, active-PR non-promotion statement, and publication boundary. Source presence is not publication evidence; live HTTPS publication and navigation remain required after the authorized Pages configuration/deployment path.
+- PR #37 is Ready at exact head `1e2f41072854edcdbaf0f9ecf14697a3bfd62195`. The head repairs the Rust 1.97.1 formatting failure in its segmented Content-Length regression; formatting, locked workspace check/test, strict Clippy, rustdoc, all 167 Python contracts, and pinned-nightly production function/line/region/branch coverage pass locally at 100%. The reader returns after the exact declared bytes, already-buffered surplus remains fail-closed, and valid self-delimited content does not require TLS EOF. Native Rust contracts and coverage have succeeded on this head, but central workflow evidence remains incomplete and no eligible approval exists.
+- PR #50 is Ready at exact head `ad87cfea59db711cb29ef90559790ba77e22029f` on protected main. At predecessor `e981ac45d0bfcd3906fc64dae5f4490edf39f9e5`, seven focused tests passed but pinned Rust formatting failed on the default-port regression. The formatter-only repair preserves all default HTTP/HTTPS and explicit-port assertions and every production source. Fresh independent verification passed all seven fresh-resolution tests and 152 Python contracts, compileall and complete Rust 1.97.1 format/check/workspace-test/strict-Clippy/rustdoc gates. Numerical coverage is 530 functions, 4509 lines, 5441 regions and 660 branches at 100%, with the unstable branch-option warning retained. Current CI `33964793228` and compatibility `33964793282` remain non-passing queue evidence. Earlier `2bd85188...` documentation and `ddbefc91...` origin-port repairs remain historical: an origin-approved IP could be paired with a different service port; the planner now binds the socket port to the effective scheme-host-port origin before I/O. ADR 0005 and doctoring retain trusted monotonic time and revalidation before socket I/O. Local admission tests do not prove an HTTP/TLS exchange, browser navigation, hosted acceptance or required approval.
+- PR #269 is Draft at exact head `7854394266d3f292e779193c01413a34f6798d7c`, stacked on #268 exact `8d4027e40b790d28d866051ba741db12927ec22c`. It adds only a fixed, product-owned `script.callFunction` text-value observation for the exact admitted current node; it performs no browser I/O and proves no post-condition. Its new documentation boundary contract and all 140 Python contracts pass locally, while hosted exact-head CI is queued and the parent-first dependency keeps it Draft.
+- PR #270 is Draft at exact head `191a14535219ea8033777fa4c970efb281b62418`, now non-force synchronized onto #269 exact `7854394266d3f292e779193c01413a34f6798d7c`. Its parent-relative transport delta is unchanged; 140 Python contracts, full Rust gates, and exact 100% local production function/line/region/branch coverage pass. Dispatch still proves neither the correlated result nor text-entry success, and fresh hosted exact-head checks plus parent-first integration remain required.
+- PR #271 is Draft at exact head `802ec806cdd4560eab48c484f435766ecabda353`, non-force synchronized onto #270 exact `191a14535219ea8033777fa4c970efb281b62418`. It admits only the exact typed observation response, treats unequal text as `PostconditionMismatch`, and retains no page-controlled or expected text in public evidence or diagnostics. Its test-first documentation contract, all 141 Python contracts, full Rust gates, and exact 100% local production coverage pass; hosted exact-head checks and ordered ancestor integration remain required.
+- PR #195 is Draft at exact head `48eb2d23009c1c804520dd5efcd0d4d072aacef1`. Its loopback regressions hold the accepted peer through opening-write timeout cleanup and locally revoked-stream classification, removing macOS close races without weakening production failures. The original affected test passed 50 consecutive focused regression passes, followed by 139 Python contracts, full Rust gates, and exact 100% local production coverage; fresh hosted exact-head checks and prerequisite integration remain required.
+- Opening-exchange fixture repair at #242 is Draft at exact head `2d0e9f69df9ade21d8e8e3d807c3ff644d83b310`, stacked on #195 exact `48eb2d23009c1c804520dd5efcd0d4d072aacef1`. A complete descendant Rust run reproduced a premature peer-close race in the mismatched-accept fixture after the existing invalid-deadline repair had been adopted. One shared test-only server now reads the complete request before replying and retains the peer through assertions in all three exchange fixtures. Fifty suite runs passed all 250 cases, followed by 139 Python contracts, complete Rust gates and exact 100% local coverage (809 functions, 7802 lines, 9959 regions, 942 branches). Production cleanup errors remain fail-closed. Native CI `33952463254` is queued; repeated local passes do not establish a zero flake rate or hosted acceptance.
+- PR #243 adopts that corrected parent by ordinary merge at `97fab641ed9d76e6c515eadcef0629edfc8064a3`, with exact base `2d0e9f69df9ade21d8e8e3d807c3ff644d83b310`. The parent-relative delta preserves the existing bounded frame transport and adds only its integration evidence; the shared fixture is identical to the owner version. All 139 Python contracts, full Rust gates and exact 100% local coverage pass (884 functions, 8733 lines, 11176 regions, 1008 branches). Native CI `33952588687` is queued, and the PR remains Draft. This repairs the previously stale #242-to-#243 dependency edge; later descendants still need ordered parent adoption and their own exact-head verification.
+- PR #246 adopts #243 at `585791f3641fbe757c3bd9fd36d5316adcc78d63`, with exact parent `97fab641ed9d76e6c515eadcef0629edfc8064a3`. Its ordinary merge preserves the message/JSON source and test blobs while carrying the owner fixture repairs. All 141 Python contracts, complete Rust gates and exact 100% pinned-nightly local coverage pass (962 functions, 9701 lines, 12427 regions, 1084 branches). Native CI `33953247053` is queued and this PR remains Draft. Fifty pre-integration fixture-suite runs passed, so no newly reproduced failure or measured failure-rate improvement is claimed. Raw protocol messages still carry no received-connection provenance or browser authority; child integration and hosted acceptance remain pending.
+- PR #247 exact head `6407895f4db4bee640074cb9c9d3cbe8b0e9e13a` merged as `b87191bcb6a95dfd7e0ed234e600639a1093c43a` into unprotected parent `feat/webdriver-bidi-text-message-assembly`; this is stack integration, not protected-main delivery. PR #248 is Draft at `b386f17c4826adabebda084bff2fba35aee94dd0`, now based on #246 `585791f3641fbe757c3bd9fd36d5316adcc78d63` by ordinary merge and explicit retargeting away from the merged #247 branch. Correlation production and Rust test blobs remain unchanged. Review reproduced that native unittest discovery collected zero checks from the free-function release contract; the same assertions now use the repository's TestCase format, collect one check and reject a missing release record. All 142 Python contracts, full Rust gates and exact 100% pinned-nightly local coverage pass (975 functions, 9848 lines, 12563 regions, 1092 branches). Native CI `33953719566` is queued. Previous Ready status and local totals did not prove enforcement of the uncollected check; current hosted verification, protected parents and received-connection authority remain unproven.
+- PR #249 is Draft at exact head `84b9407978ae0f6c115f01170b6069c601b21104`, ordinarily merged onto current #248 `b386f17c4826adabebda084bff2fba35aee94dd0`. The old `017d6e816f5a86544a63821b3ceaba94d5f17f44` tree lacked that parent and its newly discoverable release check; the parent adoption also reproduced a CHANGELOG-only merge conflict. The integration retains both release records, the parent's canonical opening fixture and native TestCase, and all four child-owned production/Rust-test blobs unchanged. All 142 Python contracts, the full Rust 1.97.1 gates and pinned-nightly 100% coverage pass locally: 989 functions, 9978 lines, 12717 regions, 1098 branches. Fresh exact-head CI `33954334610` is pending, not acceptance. The current #250 adoption is recorded next.
+- PR #250 now adopts #249 at `ec433b844a121f8554c062f92267991af9cacb6f`, with exact parent `84b9407978ae0f6c115f01170b6069c601b21104`. The conflict-free ordinary merge preserves all five child-owned production/Rust-test blobs and propagates the canonical fixture and native release TestCase. The pre-integration zero-test discovery RED becomes one executed check; all 142 Python contracts, full Rust 1.97.1 gates and pinned-nightly 100% coverage pass: 1023 functions, 10518 lines, 13525 regions, 1186 branches. Fresh exact-head CI `33955410724` is pending. Bounded status parsing still grants no authority and does not claim the later received-connection capability; current #251 adoption is recorded next.
+- PR #251 now adopts #250 at `f02af6d0dd01708d495cc08dec785675f3d58898`, ordinarily retaining both exact parents `86e8ad76838f2a64aa7e0cd56ba1f931c8d0c3dc` and `ec433b844a121f8554c062f92267991af9cacb6f`. Only the changelog required content reconciliation; both release records remain. The sender, public exports and both child-owned Rust integration tests are byte-identical to the predecessor. Native release-contract discovery first failed at zero and now executes one test; all 142 Python contracts, compileall, complete Rust 1.97.1 gates and enforced 100% coverage pass: 1037 functions, 10647 lines, 13681 regions, 1192 branches. The pinned branch-instrumentation warning below was independently reproduced on this tree too. Exact CI `33959168801` is queued; Draft, current-parent protection and hosted acceptance remain separate. A successful session-end frame write is not remote completion or process/profile teardown proof. Current #252 adoption is recorded next.
+- PR #252 now adopts #251 at `6569bf40b6595ac74c2f0a997d202137f07ba1db`, retaining predecessor `2015259529ada99af836989079cc85a15779a2d8` and current parent `f02af6d0...` by ordinary merge. The response implementation, public exports and four loopback response tests are unchanged; both conflicting release records remain. The inherited release-contract loader moves from zero RED to one executed GREEN, and all 142 Python contracts, compileall and complete Rust 1.97.1 gates pass. Enforced numerical coverage is 100%: 1044 functions, 10699 lines, 13751 regions, 1192 branches; the branch-instrumentation warning remains. Exact CI `33959789688` is queued. This response layer does not authenticate the received connection or prove resource teardown. #253's current-parent adoption is recorded next; the later #255 provenance repair remains separately owned.
+- PR #253 now adopts #252 at `afb623e4449b7cbf926fdcef7225ceaca822cfcf`, retaining predecessor `0d72082e595c0e1fcc03d609ba337896ed14e2fc` and current parent `6569bf40...` by ordinary merge. The assessment implementation, public exports and two child tests are unchanged; both conflicting release records remain. Native inherited release-contract discovery moves from zero RED to one executed GREEN. All 142 Python contracts, compileall and complete Rust 1.97.1 gates pass, with enforced numerical 100% coverage: 1054 functions, 10747 lines, 13787 regions, 1194 branches. The branch-instrumentation warning remains, and exact CI `33960670119` is queued. Review `5120013340` remains unresolved: caller-supplied teardown claims still cannot authenticate operational completion, regardless of local coverage. The #254 adoption below preserves #255's separately owned removal/provenance repair as a remaining prerequisite. This intermediate Draft is not security or operational acceptance.
+- PR #254 now adopts #253 at `b11b6c9bccd8335b58a4fb599f8ad29ac419637f`, retaining predecessor `cbaf50dcc97753cc73135497ea8225e8b18de190` by ordinary merge. The closure implementation, public exports and all six real-loopback child tests are byte-identical to that predecessor; both release records survive the changelog conflict. The inherited correlation release test moves from zero collected RED to one executed GREEN. All 142 Python contracts, compileall and complete Rust 1.97.1 gates pass, with numerical 100% coverage: 1061 functions, 10802 lines, 13843 regions, 1200 branches. Review `5120077272` remains actionable: closure kind/status still cannot establish that the acknowledged connection generation closed. Exact CI `33961562189` is queued; #255's current adoption below preserves its connection-provenance repair. Neither coverage nor loopback tests authenticate the missing connection binding, and the instrumentation warning remains.
+- PR #255 now adopts #254 at `3e7057443d7c9532ff526acb5eefe8cd4778c767`, retaining predecessor `ebac126d1632c94775c2454423575275eec45def` and parent `b11b6c9bccd8335b58a4fb599f8ad29ac419637f` by ordinary merge. All 15 owner production/test blobs are byte-identical to the predecessor. Native release-contract discovery moves from zero RED to one executed GREEN; 13 focused real-loopback tests, 142 Python contracts, compileall and the complete Rust 1.97.1 gates pass. Numerical 100% coverage is 1082 functions, 11032 lines, 14086 regions, 1202 branches. CI `33964381844` was queued at new-head observation. The previous unused-accessor repair is correctly attributed to `ebac126d...`, not its failing `63cbca0...` predecessor. Process/profile cleanup and hosted acceptance remain unproven.
+- PR #256 now adopts #255 at `ced4a851ca66c08d895a098725c7f0ad3ecf0c38`, preserving predecessor `9f2e6f29be46371762e3031a97c1cac04720694f` and current parent `3e705744...`. The core click implementation, exports and four tests are byte-identical; the correlation child delta remains only its typed click kind. Four click tests, 13 provenance/teardown tests, 142 Python contracts and complete Rust gates pass after inherited release discovery changes from zero RED to one GREEN. Numerical 100% coverage is 1088 functions, 11077 lines, 14146 regions, 1210 branches. CI `33965138401` and compatibility `33965138418` were queued; inert serialization remains distinct from authorized input or an observed browser click.
+- PR #257 now adopts #256 at `f4a8f2cbf515bea348f500b59615a9581c1b96a2`, preserving predecessor `ea2b5b78868917219c46f1304558b92490a7f6fe` and parent `ced4a851...`. The sender and two child-test files are byte-identical; the composition conflict retains both the sender and connection-bound reader. Ten focused tests, 142 Python contracts and full Rust gates pass after the inherited zero-to-one release-contract repair. Numerical 100% coverage is 1093 functions, 11131 lines, 14202 regions, 1214 branches. CI `33965440517` was queued. No focused retry was needed in this run; that does not repair the unchanged fixture's previously observed platform-level socket race. Frame-write completion is not browser-action completion.
+- PR #258 is Draft at exact head `5f830324f6d5a47ac213a57528ef95649bdfd0df`, retaining predecessor `f2ceabb3ea50b1959e936503c50cae12f3e6e480` and actual #257 `f4a8f2cb...` by ordinary merge. Its response implementation, four loopback tests and child-owned bounded socket-observation adjustment are byte-identical. Fourteen focused tests, 142 Python contracts and full Rust gates pass; numerical 100% coverage is 1100 functions, 11185 lines, 14272 regions, 1214 branches. CI `33965695405` was queued. The inherited release contract now executes. This click-response API remains protocol correlation only; the separate received-connection proof for session teardown does not automatically authenticate it or establish a browser post-condition.
+- PR #259 is Draft at exact head `66731982ed51ad62a04fcfbc759b0a33721a0254`, retaining predecessor `e1105ddf86f6c79443af8b4d306b9d34cb703c17` and current #258 `5f830324...`. Seven child production/test blobs remain byte-identical, including context admission, navigation projection and all three navigation suites. Fifteen focused tests, 142 Python contracts and full Rust gates pass after native release-contract discovery is restored. Numerical 100% coverage is 1140 functions, 11841 lines, 15129 regions, 1332 branches. CI `33965966736` and compatibility `33965966679` were queued. Exact context/URL matching is observation evidence, not transport authentication, click causation, origin rebinding or document advancement.
+- PR #260 is Draft at exact head `2c5049aff97a90958e8262b1d403bdcbd64a1e8b`, retaining predecessor `3a651967c421f77088fe25e86a63faae295390b3` and current #259 `66731982...`. The document-advance implementation, two loopback tests and Proposed ADR 0103 are byte-identical. Seventeen focused tests, 142 Python contracts and full Rust gates pass after the inherited release contract becomes executable. Numerical 100% coverage is 1154 functions, 11969 lines, 15314 regions, 1334 branches. CI `33966229736` was queued. Stale-epoch and retired-context rejection remain intact; registry advancement cannot authenticate the observation, bind the new origin or prove action causation. All six current integrations above retain the unstable branch-option warning, remain Draft and unmerged, and require their own hosted checks and policy-satisfied promotion.
+- PRs #250 through #257 remain Draft. The recorded fully locally verified chain used exact heads #250 `0eab23d5e388c5c8b984c0021a58316680c9ba8b`, #251 `86e8ad76838f2a64aa7e0cd56ba1f931c8d0c3dc`, #252 `2015259529ada99af836989079cc85a15779a2d8`, #253 `0d72082e595c0e1fcc03d609ba337896ed14e2fc`, #254 `cbaf50dcc97753cc73135497ea8225e8b18de190`, #255 `a13de5f9321e72c1867974eb7a43230f031e58df`, #256 `9f2e6f29be46371762e3031a97c1cac04720694f`, and #257 `ea2b5b78868917219c46f1304558b92490a7f6fe`. Each updated tree passed 141 Python contracts, the full Rust gates, and CI-equivalent pinned-nightly 100% production coverage. #250 and #257 each had one known macOS socket-observation race; their exact focused retries and complete coverage reruns passed. Exact-head review found that #253 still lets three unauthenticated caller booleans mint `OperationallyComplete`; raw teardown claims must remain explicitly unverified until session-bound transport, process-exit, and profile-removal evidence types are issued by their owning runtime boundaries. Hosted exact-head checks and ordered parent integration remain required. The current #255–#260 rows above supersede this older lineage only at their explicitly verified heads; no predecessor GREEN transfers to later hosted checks or protected-main delivery.
+- Reviews [#254](https://github.com/ContextualWisdomLab/OriginWeave/pull/254#pullrequestreview-5120077272) at `cbaf50dcc97753cc73135497ea8225e8b18de190` and [#255](https://github.com/ContextualWisdomLab/OriginWeave/pull/255#pullrequestreview-5120077213) at `a13de5f9321e72c1867974eb7a43230f031e58df` found a remaining connection-provenance gap. #254 passed six focused closure tests, formatting, strict network Clippy and network rustdoc locally. #255 removes the parent process/profile booleans and keeps overall teardown pending, but a new loopback regression fails when an acknowledgment from connection A is combined with closure from connection B: the assessment reports the acknowledged transport closed. Both connections can use command id 7; #254 retains only closure kind/status and the acknowledgment retains only its command id. Repair the established-stream, frame/message and command-correlation owners to preserve a non-forgeable connection-generation binding, carry it into acknowledgment and closure evidence, then reject mismatches in #255. Different-session and same-session reconnect regressions are required; endpoint equality, command-id equality and caller-supplied markers are insufficient. This is unresolved evidence integrity, not operational-completion authority or shipped behavior. Both PRs remain Draft with hosted checks queued.
+- Follow-up review `5120203295` at #255 head `92392b2da33bcd5446c6d2eb1b5504c39e60e3d6` established that received-message provenance must be checked before correlation is consumed. At that reviewed head, sender/closure generations rejected independently parsed evidence, but the two-real-socket response-substitution regression still failed: response B combined with outstanding-command state A became acknowledgment A because its generation came from the registry rather than the received text. The [exact-head review and runnable regression](https://github.com/ContextualWisdomLab/OriginWeave/pull/255#pullrequestreview-5120203295) preserve the RED lineage; the later repair evidence below supersedes this failure only for its explicitly tested snapshot.
+- Received-connection verification at #255 head `e7bfec4488b7cb4776df7b546cacb46c8c9eb13e` found that all 12 focused tests passed: the trusted reader retains one non-cloneable stream and assembler through fragments/control messages, the parser rejects a foreign generation before consuming the original command, and cross-connection closure cannot satisfy the acknowledgment. All 141 Python contracts and strict rustdoc passed. Full quality remained incomplete: formatting and Clippy failed, and pinned-nightly coverage rejected lines=11046/11049 and regions=14067/14071 even though functions and branches reached 100%. The first complete stable Rust run also exposed the inherited fixture race; its unchanged retry passed, which was not treated as repair. The fixture root cause is now repaired at #242 and integrated into #243 as recorded above. [Verification and exact uncovered paths](https://github.com/ContextualWisdomLab/OriginWeave/pull/255#issuecomment-5550208364) remain snapshot-specific; later #255 changes, hosted checks and descendant adoption need fresh verification. Overall teardown remains pending and none of this is protected-main delivery.
+- Follow-up quality verification at #255 head `63cbca0a98cf9496af981819d98029e656fc4342` completed locally at `2026-09-05T08:20Z`: 141 Python contracts, compileall and the complete Rust 1.97.1 tests passed. Formatting still failed, and strict Clippy rejected an unused private accessor left after the impossible provenance fallback was removed. Workspace check and strict rustdoc completed with that compiler warning. Pinned-nightly coverage enforcement failed at functions=1082/1083, lines=11046/11051, regions=14071/14074; branches=1202/1202. The missing-lines report identifies only the obsolete accessor's five lines; prior event/null-id and generation-exhaustion gaps are covered. The minimal owner correction is removal of that uncalled accessor plus actual pinned rustfmt, preserving received-connection validation and rejecting warning/coverage exclusions. All read-only validation handles are terminal and the existing #255 source owner has the diagnostics; no competing source change was made. Exact CI `33953476003` remains pending. These results supersede the earlier local snapshot only at this tested head, not at later repairs or protected main.
+- PR #139 at `b7ea5bfe336456fb263dd479a60b1cd0193d8a47`, based on #136 `1cffb2e23d4002f12e8462c4c8c24f404a4eeee7`, preserves all inherited documentation and limits its unique delta to the intended cleanup wrapper, regression and changelog. Both indexes retain ADRs 0007–0010 and ADR 0009 remains Proposed. All 179 Python contracts passed; [review evidence](https://github.com/ContextualWisdomLab/OriginWeave/pull/139#issuecomment-5550147096) records all three resolved findings. Current hosted Rust, coverage and pinned-Chromium runs were cancelled, so this Draft has no current hosted or browser-runtime GREEN from that verification.
+- PR #141 at `fbdf64f5818ce0c53b475196d5bcfa2ac9900846`, based on #139 `b7ea5bfe336456fb263dd479a60b1cd0193d8a47`, retains the reviewed failure/surface/timeout cleanup contracts. All 183 Python contracts passed; controlled probes confirmed that unexpected exceptions propagate after temporary-profile removal and cleanup-exit failures cannot return false cleanup evidence. The remaining two informational threads were resolved without source changes, leaving all four threads resolved; [review evidence](https://github.com/ContextualWisdomLab/OriginWeave/pull/141#issuecomment-5550364179) binds these results to the unchanged head. Hosted CI `33916583623` and pinned-Chromium run `33916583506` were cancelled. This is not current browser-runtime GREEN, adversarial erasure proof, counted approval or protected-main delivery; the PR remains Draft.
+- PR #142 at `015025f2539e4fb1dbd7d259ec22dad50f944396` passed fresh read-only failure-boundary review: 187 Python contracts passed, and controlled probes preserve unknown identity on observation failure, reject a still-live identity at the existing deadline, and accept absence or a different start-time as exit of the original identity. Both informational threads were resolved without production changes. Its exact-head CI `33916627099` and compatibility `33916627091` remain cancelled, not Linux/browser runtime GREEN.
+- PR #143 at `44fd9a450f864feff5cf2ba2883425a71ba10b9b` is a documentation and real-path regression repair on exact #142. The release-record check first failed for the missing failure-cleanup entry, then all five focused and 192 Python contracts, full Rust 1.97.1 gates and pinned-nightly 100% coverage passed: 415 functions, 3555 lines, 4444 regions, 476 branches. The production runner is unchanged. Observed exit and a surviving identity both retain failed-task status; process-observation errors leave termination unproven and record only the bounded observation-error type, losing the original browser-failure type in that fallback record. Private exception messages remain absent, and profile cleanup cannot turn a failed task into a pass. Both informational threads are resolved with this limitation documented. CI `33954715539` and compatibility `33954715568` are queued; controlled local probes do not prove real Linux/pinned-Chromium acceptance. Current #144 adoption is recorded below.
+- PR #144 at `09f2e087d0c20fe81386c18099e739a9e611a8ad` completed read-only informational review: 194 Python contracts passed, and controlled real-helper probes confirmed conservative failure when a descendant PID is reused before start-time capture, separate root and process-set deadline budgets, and the defensive exit-count invariant across all eight absent/live combinations of three descendants. Three informational threads are resolved with those limits recorded; no runtime code or timeout changed. CI `33916685294` and compatibility `33916685137` remain cancelled, not current browser acceptance. This is predecessor evidence superseded by the current adoption below; process identities appearing after sampling, cgroup ownership and OS-wide orphan absence remain outside this evidence.
+- Earlier #258 `f2ceabb3ea50b1959e936503c50cae12f3e6e480` and #259 `e1105ddf86f6c79443af8b4d306b9d34cb703c17` each passed 141 Python contracts and full Rust/numerical-coverage checks on the older #257 lineage. Their new current-parent integrations and independent 142-contract verification are recorded above; the historical measurements do not transfer.
+- PR #93 is Draft at exact head `0664f0452cb329cd692cce7f61f9001652abfda2`, based on #271 exact `802ec806cdd4560eab48c484f435766ecabda353` and synchronized with #242's complete opening-fixture cleanup. `SemanticNodeActionBinding` preserves semantic-node and business-origin identity, but does not authorize policy or execute input. Its 141 Python contracts, full Rust gates, and exact 100% local production coverage pass; hosted exact-head checks and ordered ancestor integration remain required.
+- PR #95 is Draft at exact head `97aa0f2e340ee6fd920d0418f97af276b190554f`, stacked on #93 exact `0664f0452cb329cd692cce7f61f9001652abfda2`. Only `Decision::Allow` creates policy-authorized value; other decisions fail closed, and registry-owned current authority returns `NotAdmitted` after document advance removes the node. The slice performs no browser I/O or postcondition proof. Its 142 Python contracts, full Rust gates, and exact 100% local production coverage pass; hosted exact-head checks and ordered parent integration remain required.
+- PR #96 is Draft at exact head `b7ba8dd1433410cee43084a73e31816da841b2a2`, stacked on #95 exact `97aa0f2e340ee6fd920d0418f97af276b190554f`. Registry-owned node authority is revalidated before one adapter callback, and the callback is never invoked after admission is removed. Adapter completion remains separate from postcondition proof. Its 143 Python contracts, full Rust gates, and exact 100% local production coverage pass; hosted exact-head checks and ordered parent integration remain required.
+- PR #101 is Draft at exact head `bc810b121bb0303f55afa8777a23cc0f9748c1db`, stacked on #96 exact `b7ba8dd1433410cee43084a73e31816da841b2a2`. A known-disabled interactive action fails as `NodeNotEnabled`, while `ScrollIntoView` remains selectable; retained observation state is neither current Chromium proof nor dispatch authority. Its 144 Python contracts, full Rust gates, and exact 100% local production coverage pass; hosted exact-head checks and ordered parent integration remain required.
+- PR #102 is Draft at exact head `a123c55d4839dae1db7e6671f7d4d158c7cfd9db`, stacked on #101 exact `bc810b121bb0303f55afa8777a23cc0f9748c1db`. Fresh semantic comparison rejects another node as `ObservationAuthorityMismatch`, removed action support, and newly disabled interactive state, but does not obtain or authenticate the observation or dispatch input. Its 145 Python contracts, full Rust gates, and exact 100% local production coverage pass; hosted exact-head checks and ordered parent integration remain required.
+- PR #103 is Draft at exact head `8b3416169346fa04b53c915b813d55ccf47d1876`, stacked on #102 exact `a123c55d4839dae1db7e6671f7d4d158c7cfd9db`. Same-call dispatch checks registry-owned browser authority first and fresh semantic state second; the callback is never invoked on either failure, and completion is not postcondition proof. Its 146 Python contracts, full Rust gates, and exact 100% local production coverage pass; hosted exact-head checks and ordered parent integration remain required.
+- Issue #201's release/SBOM lane remains dependency ordered, and GitHub Releases is empty. OriginWeave remains pre-GA until exact protected-head versioning, signed artifacts, SBOM/provenance, reproducibility, rollback, compatibility, security, and commercial-acceptance evidence exist.
+- The principal commercial gaps remain #27 Manifest V3/native-host isolation, #9 bounded HTTP/browser-network consumption, #10 purpose-bound protected-data runtime, #28 the first complete Chromium Agent Task vertical slice, #199 durable WARC/PROV and retention/replay, #200 stable BAP/MCP runtime API, #201 signed cross-platform distribution/update/rollback/SBOM/SLSA, #202 enterprise identity/tenant/policy/approval/audit/SLO operations, #203 exact-artifact commercial acceptance, #276 contextual-orchestrator migration, and #279 exact-head documentation verification without unnecessary Rust queue load.
+- Current evidence procedure remains fail-closed: queued reviewer evidence is non-passing; paginate the full PR inventory, bind checks/reviews/threads/workflow runs to each unchanged exact head and independently resolved live base, consult the active ruleset, and discard queued, skipped, cancelled, absent, predecessor, synthetic, status-only, and model-only evidence as passing proof.
+
+- Current PR #144 parent adoption is `3c0c5c363c2da0b4e8a621226e2f7766c735b743`, ordinarily integrating #143 `44fd9a450f864feff5cf2ba2883425a71ba10b9b`. The previous child collected three inherited failure-path contracts instead of five. Both release records and the parent's real failure-path regression now coexist with unchanged child production and process-set tests. All five focused and all 196 Python contracts, complete Rust 1.97.1 gates and pinned-nightly 100% coverage pass locally: 415 functions, 3555 lines, 4444 regions, 476 branches. Exact CI `33956094018` and compatibility `33956094011` are queued, not acceptance; the prior informational review limits remain unchanged.
+- PR #145 now adopts #144 at `bb5e8f834c37f9ce35f84db8ed1146da3659d6aa`. The ordinary merge preserves the child runner and both process-set/forced-close regressions while moving the missing inherited-contract count from three to five. All five focused and all 202 Python contracts, full Rust gates and the same four-dimension 100% coverage pass. Ten controlled actual browser-pass/trial probes distinguish observed root/set survival, observation errors and interrupted capture; session/driver cleanup alone does not prove process exit. Both informational threads are resolved with that corrected limit. Exact CI `33956596014` and compatibility `33956596013` remain queued.
+- PR #146 review repair is `812c0020bd2ecdb3eda39d999ed3327647550bdf`. Five new tests first reproduced 10 assertion failures and 7 uncaught protocol-error cases despite the predecessor's 216 passing contracts. The repair restores the non-boolean pre-shutdown count range, preserves only validated known ordinary cleanup fields, records typed terminal protocol errors without remote messages, and removes the obsolete HTTP-body close recognizer. Startup retry policy, request redaction and owned-process deadlines remain unchanged. Six new behavioral tests and all 224 Python contracts, complete Rust gates and the same 100% production coverage pass. The concurrent parent-adoption `11944410450684809ee1a71a35c77abafc5358db` is retained through an ordinary merge whose tree is identical to verified repair `d636a1829332610ada458df7b5b9d267f038a2f8`. All four review threads are resolved; exact CI `33957036553` and compatibility `33957036650` are queued, not browser acceptance. #147 must adopt this parent while retaining its own shared-deadline behavior. These three PRs remain Draft and no protected delivery, eligible approval or release is claimed.
+- Follow-up #287 review evidence at `2026-09-05T09:07Z`: Strix scan job `101243872506` executed successfully, including its quick scan and report upload. Noema job `101245089068` failed with HTTP 502 after 2739.2 seconds and one caller attempt. A fresh exact-head inventory showed no duplicate Noema run, so the same failed job was queued for a supported rerun; attempt 2 now has successful exact-head admission and queued review job `101282290732`. Queue admission is not provider recovery, executed review, a verdict or eligible approval. No caller timeout, provider, model, workflow or quality gate changed. The three central CodeQL dispatches remain queued and were not duplicated. This supersedes the earlier in-progress Noema/Strix snapshot above without transferring any old result to a new source head.
+
+- Prior #255 source-owner repair is `ebac126d1632c94775c2454423575275eec45def`, verified at that earlier owner snapshot before current-parent adoption. Its 11-file delta removes the obsolete private accessor, applies pinned formatting and repairs the Rust documentation link without removing received-connection validation. The source owner reports 141 Python contracts, full Rust 1.97.1 verification and coverage functions=1082/1082, lines=11025/11025, regions=14071/14071, branches=1202/1202. This is the owner's historical exact-tree local verification, not an independent rerun by this documentation lane or a transfer of earlier GREEN. At that observation, hosted CI `33957423557` was queued; the earlier `63cbca0...` quality RED is retained as historical RCA, and the new independent parent-adoption results are recorded above.
+- The pinned cargo-llvm-cov 0.8.6 branch report emits `warning: --branch option is unstable`. The warning was independently reproduced when reporting the existing #144/#145/#146 measurements, whose numerical verifiers still pass. Thus numerical coverage enforcement is distinct from warning-free instrumentation. The #255 source owner likewise leaves warning-free measurement acceptance outstanding. No warning suppression, coverage exclusion, denominator manipulation or alternate unreviewed measurement is used; compiler/Clippy/rustdoc results, instrumentation maturity, hosted acceptance and release readiness remain separate.
+- Central Strix #1563 advanced to `13fbb48e0b3eeca4ce7d9678add934f9bd87ad3f` from `1221b1604e1a6cfde8ca5ab7fd3e93e0fe9faf69`. The independently inspected four-file source diff adds a completed-artifact regression using the exact #166 denied-report prose and restricts console denial matching to explicit diagnostic lines while retaining warning, fatal, and workflow-error signals. This establishes a source repair, not an executed Origin acceptance result. The focused and full shell harness results plus 2890 passed, 1 skipped, and 21 subtests are owner-reported; this consumer lane did not execute them, and the skipped case is not passing evidence. At 13:46 UTC, ordinary merge `eaf9594f7fe8d8e1994349289183d6cbad056579` retained `13fbb48e...` and adopted protected central main `6f8c51d7389c22ebaf294fe8fe9ef495257883c0`. That fresh head is unmerged and hosted checks are queued; its cancelled predecessor checks and earlier local results do not validate the new merge. No Origin #166 consumer rerun was performed, so its historical executed failure remains unreplaced. Actual #219/#240 gateway and hosted execution-limit failures remain distinct.
+- Historical #147 `46e75859ef5aff2054fd6ca1720d518438e97d0f` included #146 `812c0020...` in ancestry but dropped the parent's six review-evidence regressions and reversed their associated production repairs. Native discovery collected zero tests; read-only replay of the exact parent module against that child reproduced 10 assertion failures and 8 uncaught protocol-error cases. The merge title, ancestor presence and earlier 222 passing child tests did not establish successful inheritance.
+- At predecessor #147 `af1b98ba377c73b88baa9633e2232e7f76e76f36`, the external synthesis restored parent runtime repairs and all six methods while preserving the combined forced-close observer. Its recorded RED is that all 228 native tests execute but two forced-close protocol-fault subcases fail. The inherited test injected the individual root waiter although the forced-close lane called the combined observer; a fresh focused reproduction took 10.417 seconds. The [review evidence](https://github.com/ContextualWisdomLab/OriginWeave/pull/147#issuecomment-5550841022) retains this failure and its subsequent repair rather than treating recovered ancestry as acceptance.
+- Current #147 `3dff28d9bf2dd27b72507e39979d51b8bf140fb4` repairs only that lane-specific test injection and its documentation. Ordinary cases retain the individual False result; forced-close cases inject the combined (False, False) result and assert the exact root-only identity tuple. Every existing assertion and all six methods remain; all six focused and all 228 native tests now pass, along with compileall, complete Rust 1.97.1 gates and the same 415/3555/4444/476 numerical 100% coverage. Production, deadlines, retry policy and failure denominators are unchanged; the branch warning remains explicit. Writer reassessment occurred after the temporary route closed and contacted tasks confirmed read-only status; the current repair is a normal forward push. CI `33959982049` is pending and compatibility `33959982033` is queued. The ordinary-pass two-deadline finding stays unresolved in the separate #150 scope; no hosted, protected-main or real-browser acceptance is claimed.
+- Current #148 `bded4fc9d32e5e047cea99d182aa05ae2cd03bf6` adopts that exact #147 by ordinary merge and repairs review `5120692604`: a real crash-trial startup probe first failed because the emitted browser arguments disabled Chromium's sandbox. Removing that single argument preserves the startup-failure path, driver termination and temporary-profile removal, with no unsandboxed fallback. The inherited six review regressions and both child crash-cleanup/exit-detection test blobs remain intact. The 247-test macOS run has 244 passes and three explicitly skipped Linux pidfd cases; all 247 Python contracts pass on Linux with zero skips, including a killed unreaped child, non-terminating signal and stale process identity. Complete Rust 1.97.1 gates and numerical 100% coverage pass at 415/3555/4444/476, with the branch warning retained. CI `33962062616` and compatibility `33962062608` are queued. Three other launch lanes still disable the sandbox in this branch and require their separate owner repairs; this is not runner-wide sandbox completion, pinned-Chromium acceptance, review dismissal or protected delivery.
+- Current #150 `d3c29359fd10e540ca9b3b2723bde6f94866cdb6` adopts the same #147 while preserving its existing combined ordinary-pass observer and both child test blobs. The inherited six review methods were absent before adoption. Their restoration exposed two ordinary protocol-fault failures, and full replay then exposed three sibling failure-observation cases still injecting the superseded individual waiter. Test-only reconciliation now injects paired observer outcomes and asserts the exact root-only identity tuple; incomplete process capture cannot claim full-set termination. No production timeout, retry, polling rule, success condition or denominator changed. All inherited and child checks execute, and all 229 Python contracts pass on macOS and Linux. Complete Rust gates and the same four-dimension numerical 100% coverage pass with the instrumentation warning retained. CI `33962429156` and compatibility `33962429172` are queued. The ordinary-pass deadline finding remains valid on #147 until this separate child is actually integrated; it was not resolved early.
+- Supplemental Linux verification used the existing Colima guest (Python 3.12.3, kernel 6.8.0-117-generic), without installing dependencies or changing VM configuration. A first macOS archive introduced binary AppleDouble files and caused two unrelated TLS source-decoding failures; exporting the same tracked files without filesystem metadata repaired the transport, not the tests. Final host/guest tracked-file SHA-256 manifests match exactly: #148 `c808905cbea16fd85b9519fa0a7c431d032f6692bea172950e5ba23613e7e98b`, #150 `0f5f93ed8789843ca4f228d1117328d029e3fab461f0450c75964f2ef6b98324`. These are actual Linux contract runs, not actual pinned-Chromium browser trials, hosted CI or release evidence.
+- #166 Strix run `33929688857` / job `101237371800` failed at `2026-09-05T10:53:36Z` on unchanged `e84a1a2...`. Its executed trusted central source was `a9aeee8fc94ad6002a059b380b268590ce496ef0`. The final scanner attempt exited zero, recorded completion from `10:25:17Z` to `10:53:29Z` and emitted zero SARIF results, but the console classifier also matches ordinary denied-policy prose. Replaying its exact first predicate against the sanitized final console log matches two policy-description lines and independently explains the final provider-unavailable classification. The final scanner report log has no corresponding failure-token match; the initial report-directory hypothesis was rejected. Artifact `9968177796` has SHA-256 `1ed63eb46e83d8acb61e281a6563400269a45c0924ae989ee30a4addc041048b`. Earlier attempts did have errors, so neither all-attempt success nor authoritative security acceptance is inferred. The reproduced classifier defect and exact evidence were sent to the existing central owner; no central source, provider, timeout, gate or required status was changed here, and repeating the same classifier was not used as a remedy.
+- #219 Strix run `33925442296` / job `101226800608` on `65e4315...` is a distinct failure: a real gateway HTTP 500 occurred at `08:08Z`, and the job was cancelled at `10:51:35Z` with the six-hour GitHub job execution limit explicitly recorded in its check annotation. This is neither a passing scan nor evidence that the #166 console defect caused that cancellation. Its already admitted Noema attempt 2 remains a separate queued item. Canonical provider/orchestration and runner feasibility must be reassessed before another attempt; no caller timeout or model fallback was expanded.
+- #240 Strix run `33925596890` / job `101227893828` on unchanged `24930a3...` was also cancelled at `11:20:05Z`; its annotation independently confirms the same six-hour hosted execution limit. Its actual console records gateway HTTP 500 at `07:01Z` and `10:34Z`, followed by cancellation at `11:19:59Z`. These are observed failures, not a generic inference from a red status. No successor Strix run existed in the refreshed exact-head inventory, but an unchanged retry would not repair the canonical provider or classification defects; this source/documentation lane did not start one or alter execution limits.
+
+- Temporary integration PR #291 is now closed after an external merge into `automation/147-merge-base-20260905`, whose base was #146 `812c0020...`. Live API metadata records head `52575632de07bcb90791e491b9a64b6875532ebe` and merge `24fad0498a49267982805d35d2dd57484efb4671`; Git confirms that merge is an ancestor of current #147 through forward integration `07369a37b54c11c32f810c3d5c107ad6d4b9feb4`. The temporary route neither replaced #147 nor reached protected main. The recovered runtime and test content, subsequent test-reconciliation repair and ordered acceptance requirements are evaluated separately from its merged/closed label. This lane performed no merge or closure. The fresh full inventory is 125 open / 12 Ready / 113 Draft.
+
+## Observed snapshot: 2026-08-29
+
+### Protected-main truth
+
+- Protected `main` is at `542ca1e9c0a863595b8b6697790005d2471f5413` for this snapshot. Since the 2026-08-26 observation (`b05d5acca82b9d916ada2c8e82f59f92a89817e1`), protected `main` absorbed #161 (TLS trust-bundle identifier shape). PR #170's conservative `tools/list` discovery contract is also merged into protected `main`; the complete MCP adapter remains planned.
+- Phase 0 remains complete as a reusable safety-kernel foundation: typed policy contracts, destination classification, direct TCP peer verification, TLS service identity, evidence bounds, resource mitigation, document-node authority, and protected-main tests.
+- Phase 1 is **in progress**, not shipped. The first real Chromium vertical slice still needs the active WebDriver BiDi transport stack to reach protected `main`, then compose isolated Chromium launch, session/context identity, semantic observation, typed action authorization, native browser input, post-condition proof, evidence, cancellation, crash recovery, and profile/process teardown.
+- HTTP/1.1 bounds, downloads/MIME, proxy/PAC consumption, full browser-network integration, the sensitive-data broker runtime, durable WARC/PROV capture, persistent task/API surfaces, signed cross-platform distribution, enterprise administration, and release-grade buyer acceptance remain open.
+- Active pull requests remain evidence, not shipped behavior. Successful checks on a feature or stacked branch do not prove that protected `main` contains the capability or that a child can merge before its prerequisite.
+
+### Open pull requests
+
+The live repository contained **108 open pull requests: 24 non-draft and 84 draft** when this snapshot re-paginated the complete open inventory. Compared with the prior **2026-08-28 111-PR snapshot**, stacked evidence PR #53 was merged into its unprotected feature parent; the preceding **2026-08-28 116-PR snapshot** had already recorded #71, #154, #233, #234, and #235 merging into their unprotected feature parents. PR #217 was squash-merged into the unprotected #210 feature parent, followed by PR #67 into the unprotected #64 feature parent; the current queue therefore contains no protected-main shipment. These counts are queue evidence, not protected-main delivery; protected `main` remains `542ca1e9c0a863595b8b6697790005d2471f5413`, with 11 open issues and no releases or tags. The volume and stack depth remain themselves a product-delivery risk: review, exact-head checks, dependency order, and integration truth can drift faster than a buyer-visible vertical slice reaches protected `main`.
+
+#### 2026-08-29 maintenance-loop record
+
+This snapshot re-fetched the complete open-PR inventory, the protected `main` commit, the active required-workflow ruleset, collaborator permissions, and the exact base/head pair for each representative PR. Before PR #238 was published, the same maintenance loop observed 115 open pull requests (31 ready, 84 draft); a later 116-PR recheck preceded five stack merges, leaving 111 open pull requests (27 ready, 84 draft), and subsequent child-stack merges left 108 open pull requests (24 ready, 84 draft). The active ruleset still requires one counted approving review and resolved threads; the only repository collaborator is `seonghobae`, so review provisioning remains the merge blocker for main-targeting PRs. PR #170 is merged and is not active-PR evidence.
+
+During this recheck, #71, #154, #233, #234, and #235 were merged only into their unprotected feature-parent branches after exact current-head checks and review-thread resolution. Their successful stack checks are not protected-main delivery, and their child branches retain independent evidence requirements.
+
+The sensitive-data child slice #53 was subsequently squash-merged into the unprotected #46 feature branch at merge commit `93c713a107df05385f745db4dca20091f21c4a3a`; #53 was exact head `4ecc81e59ae7bc3a640e65e2442bf30c079bd94c`. PR #46 remains an active main-targeting parent. Its current exact base/head pair is protected `main` `542ca1e9c0a863595b8b6697790005d2471f5413` to `373113119446d99f578febd39efc19366e7736b1`; the head adds the ADR 0007 predicate-boundary clarification and regression contract, with local Python/Rust verification green. Current hosted evidence remains incomplete: automatic OpenCode run `33189822385` / job `98913006386` failed closed without a current-head verdict, and central Strix run `33190794267` / job `98915422837` failed closed after three provider HTTP 500 attempts without a vulnerability report. A direct central `opencode-review` dispatch run `33192478312` / job `98921183278` was rejected because repository_dispatch actor `seonghobae` did not match configured scheduler identity `github-actions[bot]`; no qualifying non-author approval is present.
+
+The WARC/PROV child slice #217 was squash-merged into the unprotected #210 feature branch at merge commit `66f360ccac5cec60c72222cc79d58e39f6f00088`; #217 was exact head `6b8a3fdeae52ad94b90086bbc9b42863b90c9614`. PR #210 subsequently advanced to current exact head `7946dce9a3dd074047d93fca299d48c7aef40e47` after its merged-child attribution repair, recursively encoded-control repair, and exact coverage repair; this stack remains active-PR evidence and not protected-main delivery or approval evidence.
+
+The browser-task interruption child slice #67 was squash-merged into the unprotected #64 feature branch at merge commit `5021d142583cb5a8e393248048bb824762a98056` from exact PR head `25ab76e8279d4a904d04afeb264bac3e89f46b45`. PR #64 consequently advanced from `debc761aa59aee1509b7a260474fa33216453511` to current exact head `5021d142583cb5a8e393248048bb824762a98056`; its exact-head hosted checks were regenerating at this snapshot, with no unresolved inline review threads. This stack remains active-PR evidence and not protected-main delivery or approval evidence.
+
+The later exact-head recheck also recorded the WARC/PROV retention-lifecycle slice: #239 is Draft at `e840ca299d29a15223c8b9bb1397002c4f41b4a3` on #227 head `e45cd6cdcdee73b5c16dc942e6c98cb7e745fae0`, and #237 is Draft at `2459af602e72fbfe1ce816919473a1075ec0c41f` on protected `main`; their current exact checks and reviews remain independently actionable evidence. Neither active branch is protected-main behavior.
+
+The same exact-head recheck corrected workflow provenance: the active ruleset's seven required workflow entries (`close-empty-pr`, `opencode-review`, `pr-review-merge-scheduler`, `security-scan`, `strix`, `sast-semgrep`, and `noema-review`) point to the central `.github` repository, repository ID `1274066402`, where all seven files exist on `main`; their absence from the OriginWeave tree did not prove missing workflow identities. On PR #210 head `0341079331f9cea669eb9a5cc21842fd6027431e`, run `33177641855` failed closed because no OpenCode current-head verdict existed, while run `33177641888` failed closed after three bounded attempts because the Strix provider/backend returned an internal server error and produced no vulnerability report. A `gh run view` workflow-endpoint 404 for the external workflow IDs is a lookup mismatch, not passing evidence or proof that the identities are absent. A bounded central scheduler dispatch was accepted as run `33178984025` with auto-merge and branch updates disabled; no result is transferred until the exact head is revalidated. These failures block affected PRs until current-head review and security evidence complete. This does not authorize bypass, self-approval, stale checks, or weaker gates.
+
+The WARC resource-record slice #210 is now Ready; PR #210 current exact head is `7946dce9a3dd074047d93fca299d48c7aef40e47` based directly on protected `main` `542ca1e9c0a863595b8b6697790005d2471f5413`, after incorporating #217, correcting the merged-child attribution, and closing the recursively encoded-query-control coverage gap. Its predecessor head `5f59947f5e4b0d3bc0aa5b2d4c6722d3b7c43047`, prior stack merge head `66f360ccac5cec60c72222cc79d58e39f6f00088`, earlier exact head `bea65643109449d63d367a35b8d9bf327ee7cb2c`, and their OpenCode/Strix provider failures remain historical evidence only. At the current head, `Rust contracts` job `98942518975` and `Production coverage` job `98942518680` succeeded, while `noema-review` job `98942513421` and `strix` job `98942803402` remain in progress; the current-head OpenCode verdict is still absent and no counted approval exists. Central repair PR #1391 was opened at historical head `e4ba6b599cd1e50d0139762885682607b731655d` and is now open at exact head `36ac3aa71b2580685f84d416a81e42c39dee927c` on current central `main` `e1b03eebc6dc5c85aed393e5928927c96376cf46`. Its branch-update merge and follow-up prompt hardening are not approval or coverage evidence.
+
+The documentation refresh PR #238 remains Ready on protected `main` `542ca1e9c0a863595b8b6697790005d2471f5413`; its moving current head is intentionally authoritative in live GitHub metadata and the PR body rather than repeated in this self-referential baseline, while local full-suite evidence is green. The immediately preceding PR #238 head `d0b0d1ed92f891f14646fc673b8e1c0d912586fd` remains historical: automatic OpenCode run `33193822920` / job `98926243116` failed closed without a current-head verdict, current Strix run `33193822929` / job `98925769697` succeeded, and central dispatch run `33194506918` / job `98928580387` also failed closed at OpenCode after its validator, bootstrap, and coverage jobs succeeded. The preceding exact-head OpenCode run `33184553025` / job `98894986761` and earlier targeted run `33182749298` remain historical; targeted run `33182749298` dispatched OpenCode run `33182772296`, which failed closed as `MODEL_OUTPUT_UNAVAILABLE` with `model pool exhausted`, and optional cross-repository status publication was denied with HTTP 403. The current Devin documentation-evidence finding requiring an enforcing `tools/list` traceability contract has been implemented and its review thread resolved. No non-author counted approval exists, so these are review-tool/documentation findings, not approval or protected-main shipping evidence.
+
+The controlled Chromium prerequisite stack was also re-fetched after its exact-head repairs: #70 is Ready at `441a8ce1d09c329c5c1168f4906d9a38fd0abc01` on protected `main` `542ca1e9c0a863595b8b6697790005d2471f5413`; #71 is merged into the unprotected #70 feature branch; #72 is Draft at `600d3975c02b68da1974a4c73069b966b39dce7b` on the retained #71 branch; and #73 is Draft at `ce1b138509ab4f52cb0f80290f104358473c6ed3` on #72. #70's Rust, coverage, pinned-Chrome, and ordinary security checks are successful, while exact-head OpenCode failed closed without a current verdict and exact-head Strix failed closed after three provider HTTP 500 attempts; all current Devin informational threads are resolved. #72 and #73 retain independent exact-head evidence requirements. #82 is Ready at `f5776f5f233ac0a7c05e3f4a2846436c23438043` on protected `main`; its Rust, coverage, Chrome, and ordinary security checks pass, exact-head OpenCode failed closed without a current verdict, and its current Devin informational thread is resolved. #152 is Ready at `81407a0e5189a413d1be0963fea90a0c2f254ce1` on protected `main`; its source, coverage, and security checks are successful except exact-head `opencode-review`, which failed closed without a current-head verdict. These are active-stack evidence only, not protected-main behavior or merge authorization.
+
+#### Current exact-head active PR evidence
+
+The following representative slices were re-fetched from GitHub for this snapshot. Their exact base/head pairs are recorded so later checks, reviews, and restacks cannot be confused with predecessor evidence:
+
+| PR | State | Exact base head | Exact head |
+|---|---|---|---|
+| #73 | Draft | `600d3975c02b68da1974a4c73069b966b39dce7b` | `ce1b138509ab4f52cb0f80290f104358473c6ed3` |
+| #72 | Draft | `f86ce504138e79d6e95141a441f60b40920e1fa6` | `600d3975c02b68da1974a4c73069b966b39dce7b` |
+| #46 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `373113119446d99f578febd39efc19366e7736b1` |
+| #70 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `441a8ce1d09c329c5c1168f4906d9a38fd0abc01` |
+| #82 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `f5776f5f233ac0a7c05e3f4a2846436c23438043` |
+| #210 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `7946dce9a3dd074047d93fca299d48c7aef40e47` |
+| #64 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `5021d142583cb5a8e393248048bb824762a98056` |
+| #237 | Draft | `542ca1e9c0a863595b8b6697790005d2471f5413` | `2459af602e72fbfe1ce816919473a1075ec0c41f` |
+| #239 | Draft | `e45cd6cdcdee73b5c16dc942e6c98cb7e745fae0` | `e840ca299d29a15223c8b9bb1397002c4f41b4a3` |
+| #229 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `0145ccba5901e301b41d4be674ca1ed23483ad37` |
+| #220 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `b11db2be68f9b6d71aa4c4290b97a8b22097b353` |
+| #211 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `52a918577958a5701e1146c7eb8b62fe8f8ccd44` |
+| #152 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `81407a0e5189a413d1be0963fea90a0c2f254ce1` |
+| #195 | Draft | `6922dd98779e8f8aad132a3b1f563d7ba6e6d070` | `48eb2d23009c1c804520dd5efcd0d4d072aacef1` |
+| #242 | Draft | `48eb2d23009c1c804520dd5efcd0d4d072aacef1` | `55fef0c3fae1724eddada53e52c4a0311f509aa3` |
+| #248 | Ready | `6407895f4db4bee640074cb9c9d3cbe8b0e9e13a` | `7d6db16b2ead201fcec320854923f90d3ad0d8bc` |
+| #249 | Draft | `de7754aaeb97ccb0fd47bcbe1c4d99c10eaf84eb` | `2279d18189fcd6cdb2b38aca53b877434d41c913` |
+| #250 | Draft | `2279d18189fcd6cdb2b38aca53b877434d41c913` | `cbddf507ac41080ee65230a8d6047dd8d06fd719` |
+| #251 | Draft | `cbddf507ac41080ee65230a8d6047dd8d06fd719` | `99e2eb946e7ebbffa68f65a00d25243a2cf4242a` |
+| #252 | Draft | `99e2eb946e7ebbffa68f65a00d25243a2cf4242a` | `881a599fb3a920b6bfd4f0a276f3cf24a61d8194` |
+| #253 | Draft | `881a599fb3a920b6bfd4f0a276f3cf24a61d8194` | `5e2b17c7a8d1953bb06a41e0296f801f0d015a9a` |
+| #254 | Draft | `5e2b17c7a8d1953bb06a41e0296f801f0d015a9a` | `a4841016b94cb18e917d250a9ca9149af54ceef0` |
+| #255 | Draft | `a4841016b94cb18e917d250a9ca9149af54ceef0` | `f8edec38cf8ab7fde22b8d1de9305728c1a2f25b` |
+| #256 | Draft | `f8edec38cf8ab7fde22b8d1de9305728c1a2f25b` | `bd1f5ac60a76d2edb35e63095d406e53bc43931f` |
+| #257 | Draft | `bd1f5ac60a76d2edb35e63095d406e53bc43931f` | `ac73abfe7edd5786a7eb3eaab1a8c773093be7d3` |
+| #93 | Draft | `802ec806cdd4560eab48c484f435766ecabda353` | `0664f0452cb329cd692cce7f61f9001652abfda2` |
+| #95 | Draft | `0664f0452cb329cd692cce7f61f9001652abfda2` | `97aa0f2e340ee6fd920d0418f97af276b190554f` |
+| #96 | Draft | `97aa0f2e340ee6fd920d0418f97af276b190554f` | `b7ba8dd1433410cee43084a73e31816da841b2a2` |
+| #101 | Draft | `b7ba8dd1433410cee43084a73e31816da841b2a2` | `bc810b121bb0303f55afa8777a23cc0f9748c1db` |
+| #102 | Draft | `bc810b121bb0303f55afa8777a23cc0f9748c1db` | `a123c55d4839dae1db7e6671f7d4d158c7cfd9db` |
+| #103 | Draft | `a123c55d4839dae1db7e6671f7d4d158c7cfd9db` | `8b3416169346fa04b53c915b813d55ccf47d1876` |
+| #124 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `fdb88698ca20626a6643bc2ad7944fb968835700` |
+| #37 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `5e3dfcbd7a4daea297782cb99635990368589232` |
+
+These rows are delivery evidence only. None has counted independent approval in the current collaborator inventory.
+
+The current documentation branch is PR #238 itself, so its self-referential exact-head row is intentionally omitted; GitHub PR metadata and the PR body are the authoritative current-head record for this change.
+
+#### Historical 2026-08-26 maintenance-loop record
+
+The interactive maintenance loop performed the following verified state changes on exact heads; none of them is protected-main behavior until merged:
+
+| Action | Exact evidence |
+|---|---|
+| Supersession closure | #153 closed with replacement evidence: base-stack tip (`4da223ac`) already implements `_terminate_owned_process_bounded` exit-race tolerance that supersedes the branch delta |
+| Conflict reconciliation | Merge commits pushed to #37 (`27f6acd6`, ci.yml aligned to reviewed `nightly-2026-08-18` pin), #149 (`7852a540` + rustfmt fix `54f96008`), #152 (`65b0c705`), #173 (`ecc9574a`), #175 (`765c88f6`, keeps `crate_root.rs` naming) |
+| Governance remediation (#212) | #43 reconciled with main in `04e262d5`; the `chrome_sandbox` workflow mutation was first removed, then restored under recorded independent authorization (issue #212 option (b)) because the PR's own contract test fails closed without it; fresh exact-head checks re-ran on the restored head |
+| Security finding fix (#124) | Strix vuln-0001 (Unicode homoglyph path confusion, MEDIUM) remediated in `30cc458b`: audited workflow paths now restricted to a canonical ASCII alphabet with homoglyph/fraction-slash/fullwidth regression contract tests; CHANGELOG updated |
+| Fail-closed provider re-dispatch | ~21 failed Strix required-check runs re-dispatched on unchanged exact heads; completed reruns returned success on #46, #48, #156, #157, #159, #218, and #219 heads at snapshot time; cancellations only where newer heads superseded the run |
+| Current-head review re-dispatch | Central merge-scheduler dispatches sent for #47, #62, #63, #65, #74, #166, #173, #175, and #220 because their stale `CHANGES_REQUESTED` verdicts cited coverage-evidence results that are green on the same heads today |
+
+#### Organization review-pipeline congestion record
+
+Between 2026-08-26T02:44Z and 2026-08-26T03:35Z the organization-wide Actions queue exhibited a systemic backlog: scheduler, OpenCode-review-dispatch, Noema, and Strix runs across `.github`, `naruon`, `pg-erd-cloud`, and OriginWeave sat `queued`/`pending` while only single-digit runs were `in_progress`. This delays every current-head AI review and therefore every ruleset-gated merge. It is an infrastructure-capacity signal, not a code defect, and it does not authorize merging without current-head review evidence.
+
+Representative active workstreams at this snapshot were:
+
+| Workstream | Representative active PR evidence | Delivery boundary |
+|---|---|---|
+| Product baseline | (merged: #196 on 2026-08-24) | Baseline publication reached protected `main`; this document is its successor snapshot |
+| Presentation identity | #229 at `fb868589d065c2cea0b9c8c0f5e655a89f42bee6` onto `542ca1e9c0a863595b8b6697790005d2471f5413` | Ready/non-draft local privacy kernel; current required checks include a failed Strix run, and the PR remains blocked without counted approval; no protected-main shipment is claimed |
+| Enterprise approval authority | #220 at `b11db2be68f9b6d71aa4c4290b97a8b22097b353` onto protected `main` `542ca1e9c0a863595b8b6697790005d2471f5413` | Ready/non-draft bounded maker-checker approval lifecycle on the exact `ApprovalScope`; current checks and review state require a fresh exact-head audit, with no counted approval |
+| Release artifact identity | #218 and #219 | Ready/non-draft fail-closed benchmark release decision and canonical release manifest binding; Strix provider-failure reruns completed green on both heads |
+| Schema-bound extraction and BAP lifecycle | #209 and #208 | Ready/non-draft schema-bound extraction contract and resumable task-lifecycle kernel; #209 Strix rerun green, #208 rerun re-dispatched after a further provider failure |
+| WebDriver BiDi transport | #188 through #205 | Active stack whose top #205 merged into its prerequisite branch, not protected `main`; it exercises framed `locateNodes` exchange over a bounded WebSocket opening path, but authenticated browser-process provenance, semantic task execution, and protected-main shipment remain unproven |
+| MCP adapter | (#168 and #170 merged) | Typed MCP routing and conservative `tools/list` cache metadata are protected-main behavior; the complete MCP transport, OAuth, browser I/O, and persistence adapter remains planned |
+| Workflow-registry audit | #124 | Real Strix finding vuln-0001 (Unicode homoglyph path confusion, MEDIUM) remediated on head `30cc458b` with regression contract tests; fresh exact-head checks and review re-running |
+| Controlled Chromium and recovery | #65, #70, #72-#73, #100, #105, #142-#152 and descendants | Real pinned-browser fixture, semantic location, resource, crash, and teardown evidence exists on active stacks; #71 is merged only into the #70 feature branch, and evidence does not transfer across heads or prerequisites |
+| Durable WARC/PROV evidence | #210, #217, #239 | Bounded WARC resource records, PROV JSON-LD binding, and retention-lifecycle boundaries are active-PR foundations; durable ownership, replay, retention/deletion, and browser side-effect reconciliation remain open |
+| Manifest V3 and native messaging | #27, #43 governance remediation, and the extension/native-host stack including merged #154 and active #169 | Compatibility and Agent-authority isolation remain incomplete until exact release artifacts and platform matrices are proven; #43's sandbox workflow mutation is now owner-authorized under issue #212 option (b) |
+| Sensitive-data and model route policy | #10 and its active policy stacks | Deterministic policy values exist, but trusted broker execution, retention/deletion, runtime isolation, and auditable product workflows remain open |
+| VPN/profile intent | #149 | Bounded WireGuard/IKEv2 profile authority reconciled with main (`54f96008`); it does not create a tunnel, route, DNS state, authenticated gateway, or connectivity proof |
+
+PR #205 head `f427aa69151987d7e3369bd96d5739ea38d0f7ad` merged as `6c5ef5e2079d54c617183ecfa757e406f48f0aea` into stacked prerequisite branch `feat/webdriver-bidi-websocket-frame-transport` at base `c1bc7e78f3a9debf4f517fb6b5f11dd67be4ad92`. Its successful exact-head checks are stacked-branch integration evidence only; the current protected `main` is `542ca1e9c0a863595b8b6697790005d2471f5413`.
+
+#### Historical exact-head active PR evidence: 2026-08-26
+
+The following newest slices were re-fetched from GitHub for this snapshot. Their exact base/head pairs are recorded so later checks, reviews, and restacks cannot be confused with predecessor evidence:
+
+| PR | State | Exact base head | Exact head |
+|---|---|---|---|
+| #220 | Ready | `b05d5acca82b9d916ada2c8e82f59f92a89817e1` | `e0740a6f3a41067a4460249378e0266815018a74` |
+| #219 | Ready | `b05d5acca82b9d916ada2c8e82f59f92a89817e1` | `3e34a54ae279686a28309d59b8b3b9bfbd283a80` |
+| #218 | Ready | `b05d5acca82b9d916ada2c8e82f59f92a89817e1` | `911ea33d8a5aca7673307bb6fdcad4b450f5c111` |
+| #209 | Ready | `b05d5acca82b9d916ada2c8e82f59f92a89817e1` | `b35d739017aa5d361b605be48045be50b5a35f6f` |
+| #208 | Ready | `b05d5acca82b9d916ada2c8e82f59f92a89817e1` | `e41d3be4c290c4e434aac33d777e511dfb94e03d` |
+| #124 | Ready | `b05d5acca82b9d916ada2c8e82f59f92a89817e1` | `296ad25bb541023dbc869ae07ae1d853820f83a4` |
+
+These rows are delivery evidence only. None has counted independent approval in the current collaborator inventory, and predecessor rows from earlier snapshots are retained below as regression anchors that must never be promoted to current-head evidence.
+
+#### Historical regression-anchor exact-head evidence: superseded 2026-08-24 rows
+
+The following rows were current on 2026-08-24 and are retained only as regression anchors; every listed head has since been superseded or merged and must never be promoted to current-head evidence:
+
+| PR | State | Exact base head | Exact head |
+|---|---|---|---|
+| #222 | Draft | `56fcfa56525e4f2e980e0ee05b6776d621bcddc5` | `1e2ce3d4071a1a75ee891bdcd71c506b3b50d4bc` |
+| #221 | Draft | `8145d40f1b028a8f4dc7e7da47ac89bb9e5bb2c7` | `6f339df1e5b3ddb265f4ddd7b262d4de1e0b5e1f` |
+| #220 | Ready | `0841d2ab3d8b5e60a03c0a8e818cf438e2716829` | `ed4cab16cf88c76ce1c145a22d0a274ef2d57263` |
+| #219 | Ready | `0841d2ab3d8b5e60a03c0a8e818cf438e2716829` | `8145d40f1b028a8f4dc7e7da47ac89bb9e5bb2c7` |
+| #218 | Ready | `0841d2ab3d8b5e60a03c0a8e818cf438e2716829` | `49e98fba6974219b3bb0336c822b12667f1e1c03` |
+| #216 | Ready | `0841d2ab3d8b5e60a03c0a8e818cf438e2716829` | `75130851a0f7ce528a7a36382eb026ac7942a0aa` |
+| #214 | Draft | `40d642d5470a7753b8211907c190367f742f2f12` | `f79999681866ecf0e5fe17d895170f3f6cae7361` |
+| #211 | Draft | `85cc477688246900697f4cfb91c0c8f1f692934a` | `40d642d5470a7753b8211907c190367f742f2f12` |
+| #210 | Draft | `c38b9665774d6b3754e572bed527737b5e179833` | `529d11a3571f6b1834b9baa49ef67eb08f043978` |
+| #209 | Ready | `0841d2ab3d8b5e60a03c0a8e818cf438e2716829` | `c38b9665774d6b3754e572bed527737b5e179833` |
+| #208 | Ready | `0841d2ab3d8b5e60a03c0a8e818cf438e2716829` | `85cc477688246900697f4cfb91c0c8f1f692934a` |
+
+The stack topology shows #209 → #210 → #217 → #222 (WARC/PROV chain, with #217 merged into #210's unprotected branch), #208 → #211 → #214 (BAP chain), and #218 → #221 → #220 (release/enterprise chain) at this snapshot. Every active row above remains PR evidence; none is protected-main behavior.
+
+### Required-check provider failure record
+
+On 2026-08-23 the required Strix security scan failed closed on exact heads of #220 (`ed4cab16…`), #218 (`49e98fba…`), and #208 (`85cc4776…`) because its LLM provider/backend was unavailable (rate limit, token cap, connection, warm-up, or model-behavior failure); no vulnerability report artifact was produced, so the workflow correctly refused to convert an incomplete scan into passing security evidence. Failed jobs were re-dispatched on the unchanged exact heads on 2026-08-24 and again on 2026-08-26. This is a provider-infrastructure failure record, not a weakening of the fail-closed gate or a substitute for a completed authoritative scan.
+
+On 2026-08-26 rerun outcomes were verified per run: completed reruns returned `success` on the heads of #46, #48, #156, #157, #159, #218, and #219; several earlier runs for #37, #43, and #149 were cancelled only because conflict-reconciliation pushes created newer heads with fresh scans; remaining reruns were still in flight at snapshot time. One rerun (#124) produced a real MEDIUM finding (vuln-0001) instead of provider noise; that finding was remediated on the branch head rather than suppressed, preserving the fail-closed contract.
+
+#### #195/#198 WebDriver BiDi opening path status
+
+Phase 1 is **in progress**, not shipped. #195 and #198 provide bounded WebSocket opening-path evidence on active branches; framed BiDi commands, authenticated browser-process provenance, semantic task execution, and protected-main integration remain open.
+
+#### #149 VPN/profile intent status
+
+PR #149 is a ready (non-draft) pull request whose conflict reconciliation and rustfmt correction landed on head `54f96008` on 2026-08-26; it still only describes bounded WireGuard/IKEv2 profile authority and does not create a tunnel, route, DNS state, authenticated gateway, or connectivity proof.
+
+The current queue must be processed in dependency order. A green child branch cannot substitute for current checks and review on its prerequisite, synthetic merge, or eventual protected-main commit. PRs that only duplicate, supersede, or preserve stale branch topology should be closed with explicit replacement evidence rather than retained indefinitely; this loop exercised that policy by closing superseded #153 with replacement evidence.
+
+### Review and merge authority
+
+The active `CWL Central required workflows` ruleset (re-fetched for this snapshot) requires one approving review, resolved review threads, no last-push approval requirement, `merge`/`squash` merge methods, and seven configured required workflows (`close-empty-pr`, `opencode-review`, `pr-review-merge-scheduler`, `security-scan`, `strix`, `sast-semgrep`, and `noema-review`). The current collaborator inventory contains only `seonghobae` with administration and push permissions, creating a **reviewer-provisioning gap** for counted non-author approval.
+
+This gap does not authorize self-approval, stale-head merges, administrative bypass, or weaker checks. Because the current GitHub ruleset independently requires a counted approval, the solo-maintainer hold does not satisfy the live merge gate: an eligible non-author collaborator must submit a formal `APPROVED` review on the current head. Until that reviewer-provisioning gap is repaired, protected-main merges stop even when exact-head checks, security gates, complete coverage, rustdoc/Clippy, threads, and AI-review evidence are otherwise complete. Before any merge decision, re-fetch the exact ruleset, collaborators, PR head/base, reviews, unresolved threads, and required checks; do not assume this dated observation remains current.
+
+### Open issues and governance signals
+
+This snapshot contains 11 open issues plus 2 governance signals, for the 13 rows below. Governance signals remain visible because they affect delivery authority but are not counted as product or operational issues.
+
+| Issue or signal | Current gap or signal |
+|---|---|
+| #28 | First real Chromium Agent Task vertical slice; highest immediate Phase 1 buyer-visible gap |
+| #27 | Complete Manifest V3 compatibility and extension-authority isolation matrix |
+| #9 | Bounded HTTP/1.1 semantics over the authenticated TLS stream |
+| #10 | Purpose-bound operational PII disclosure and trusted broker/storage lifecycle |
+| #123 | Fleet incident: disable orphaned TLS, HTTP, and one-shot workflow identities |
+| #187 | Manual-authority review of the coverage-diagnostics workflow delta |
+| #212 | Governance: remove or independently authorize the PR #43 MV3 workflow mutation — **option (b) executed 2026-08-26** with owner-directed authorization recorded on the issue and the mutation restored on the reconciled branch; re-evaluate if the authorization record is contested |
+| #215 | Governance: restore an enforceable protected-main policy that does not create a routine admin bypass |
+| #199 | Schema-bound extraction with durable WARC/PROV replay, retention, deletion, and offline verification |
+| #200 | Stable BAP/MCP runtime API with authenticated, idempotent, cancellable, resumable task lifecycle |
+| #201 | Signed cross-platform Chromium distribution, installer/updater, patch SLA, rollback, SBOM, and provenance |
+| #202 | Enterprise control and experience plane: operator UI, Keyverse-compatible identity, tenancy, approval, audit, SLO, Figma, and Storybook |
+| #203 | Release-grade web-agent benchmark and commercial acceptance gate bound to exact signed artifacts |
+
+Issue #206 (harden-runner custom detection initialization failure) was closed after its remediation landed on protected `main` between snapshots.
+
+The five newly separated product-completion tracks are **durable WARC/PROV replay**, **stable BAP/MCP runtime API**, **signed cross-platform Chromium distribution**, **enterprise control and experience plane**, and the **commercial acceptance gate**. They are separate issues because each has a distinct authority, data, release, and buyer-acceptance boundary.
+
+The hourly product-development loop is operational infrastructure, not proof that a browser product, issue, pull request, or release meets buyer acceptance.
+
+## Buyer-visible and technical gap matrix
+
+| Priority | Buyer-visible outcome | Protected-main status | Completion issue and acceptance evidence |
+|---|---|---|---|
+| P0 | A bounded task observes a real Chromium page, performs one typed action, verifies the post-condition, and emits provenance | **Open / Phase 1** | #28; repeated real Chromium E2E with isolated context, exact session/node authority, typed dispatch, post-condition, crash cleanup, and protected-main checks |
+| P0 | Navigation consumes approved origin, resolution, route, TCP peer, TLS identity, bounded HTTP, redirect, MIME, and download policy | **Partial foundation** | #9 plus #28; real browser-network adapter proves the governed path is consumed end to end |
+| P1 | Existing Chromium extensions remain compatible while Agent authority stays separate | **Partial active-PR evidence** | #27; exact supported-build/platform compatibility matrix, managed allow-list, native-host isolation, repeatability, and release binding |
+| P1 | Authorized work can use necessary PII without ambient exposure | **Policy foundation; runtime open** | #10; opaque broker, exact field/purpose/destination/model policy, atomic use/revocation, retention/deletion, and value-free telemetry |
+| P1 | Every released structured field is traceable to replayable source evidence | **Foundations only** | #199; durable WARC/PROV replay, integrity, retention, deletion, offline verification, extraction precision/recall, and 100% provenance completeness |
+| P1 | External Agents integrate through a stable, authenticated product contract | **Partial active-PR MCP primitives** | #200; BAP 1.0, MCP 2026-07-28 adapter, idempotency, task cancellation/resume, checkpoint/reconciliation, and SDK conformance |
+| P1 | Buyers can install, update, verify, and roll back a supported product | **Not shipped** | #201; signed Windows/macOS/Linux/headless artifacts, Chromium revision manifest, updater security, patch SLA, SBOM, SLSA provenance, and recovery |
+| P1 | Enterprise teams can provision, approve, audit, operate, and recover the service | **Not shipped** | #202; Keyverse-compatible OIDC/SCIM, tenant isolation, policy/approval/evidence UI, SLO/incident controls, data residency, CSAP/SOC 2 evidence mapping, WCAG 2.2, Figma File ID, and Storybook |
+| P0 | A release has reproducible proof of usefulness, safety, evidence completeness, and recovery | **No product-wide release gate** | #203; deterministic, compatibility, adversarial, recovery, and enterprise suites with statistical reporting and an exact-artifact commercial acceptance gate |
+| P0 | Valid changes reach protected `main` without authority improvisation or unbounded stack growth | **Blocked / high integration debt** | Shrink the open-PR queue in dependency order, provision legitimate review authority, require exact-current evidence, and close duplicates/superseded branches |
+
+## Commercial completion definition
+
+OriginWeave is not complete merely because every low-level primitive exists in some open branch. A release candidate is commercially complete only when all of the following are true for the declared support profile:
+
+1. #9, #10, #27, and #28 are integrated on protected `main` as a complete browser/network/action/evidence chain.
+2. #199 provides replayable, retention-governed evidence for every released structured result.
+3. #200 exposes a stable authenticated runtime API and task lifecycle without raw Chromium authority leakage.
+4. #201 produces signed, updateable, rollback-capable release artifacts bound to Chromium, SBOM, and provenance.
+5. #202 supplies tenant-safe enterprise administration, approvals, audit, SLOs, incident recovery, accessible Figma/Storybook-backed UX, and control evidence.
+6. #203 accepts the exact signed artifacts through a reproducible benchmark; missing or inconclusive evidence cannot be promoted to success.
+7. Production function, line, region, and branch coverage and public API documentation remain exactly complete for OriginWeave-owned code.
+8. CHANGELOG, version, supported-platform matrix, security policy, runbooks, licensing, release notes, upgrade/rollback guidance, and procurement evidence match the exact release.
+9. No required check, browser/platform lane, security case, benchmark case, or independent review is skipped, stale, inherited, or represented by status-only evidence.
+10. The open PR queue is reduced to bounded active work rather than being the only place where the product exists.
+
+## Historical next executable queue
+
+1. Drain the merge gate in dependency order: for every ready root PR whose current head is check-green with resolved threads, obtain the current ruleset's counted `APPROVED` review from an eligible non-author collaborator; OpenCode approval or skip evidence does not substitute for that GitHub review. If no eligible approver exists, record the reviewer-provisioning gap and do not merge. Root candidates include #37, #40, #43, #45–#48, #51, #62–#65, #74, #82, #124, #149, #152, #156–#166, #173, #175, #208, #209, #211, #218, #219, #229, #237, #238, and #239 as their current checks land. Treat dependent children separately: only after a predecessor reaches protected `main`, retarget and independently revalidate its immediate child; preserve orders such as #218 → #221 → #220 rather than treating #208–#220 as a flat merge range.
+2. Keep the organization review pipeline healthy: monitor the central Actions backlog recorded above; if OpenCode reviews stop landing on OriginWeave heads while the queue is idle, repair `ContextualWisdomLab/.github` dispatch/concurrency configuration rather than weakening any gate.
+3. Finish the #9/#28 browser-network and Chromium vertical slice, including the #181–#205 WebSocket opening path and framed BiDi command/response stack, then semantic observation, policy, action, post-condition, and recovery boundaries on protected `main`.
+4. Finish #27 and #10 as separate security tracks; neither should be hidden inside the first browser PR.
+5. Implement #199, then #200, so durable evidence and stable task authority precede broad enterprise integrations.
+6. Implement #201 before making release/support claims; exact CI browser evidence must be bound to the actual signed artifact.
+7. Design #202 in Figma, record the Figma File ID in the ADR, implement reusable design tokens and Storybook components, then add identity/tenant/approval/audit/operations integration.
+8. Make #203 the final release gate across the exact signed distribution, not a source branch or model narrative.
+9. Only after the commercial acceptance gate passes, increment the version, finalize CHANGELOG/release notes, publish signed artifacts, and verify upgrade/rollback from the prior supported release.
+
+## Evidence commands
+
+Run `scripts/ci/collect_live_merge_evidence.sh` to collect reusable, exact-head and exact-base merge evidence before interpreting the volatile inventory below.
+
+The executable [collect_live_merge_evidence.sh](../scripts/ci/collect_live_merge_evidence.sh) procedure paginates the complete open-PR and open-issue inventories, excludes pull requests from the issue count, and binds each merge verdict to the exact head, base, checks, reviews, and review threads. It is the canonical procedure; do not copy it into this baseline.
+
+The branch-scoped rules response determines the active rules affecting `main`; each PR's exact `HEAD_SHA` then determines which check runs, legacy statuses, workflow runs, reviews, and unresolved threads are current. The saved merge verdict binds counted approvals to the latest review per eligible collaborator, excludes the PR author, and requires `APPROVED` on the exact head. It deliberately does **not** infer GitHub's actual last-push actor from commit author or committer metadata: when `require_last_push_approval` is active, this portable evidence procedure records `github_rule_evaluation_required` and keeps `approval_gate_satisfied` false until GitHub's authoritative rule evaluation is consulted. The saved PR JSON also preserves the exact base reference and branch ancestry input for the dependency graph. Evidence is retained only when both `RECHECKED_HEAD_SHA` and `RECHECKED_BASE_SHA` match the collected values; a moving head or base discards the temporary verdict, and three failed attempts leave no unstable merge verdict.
+
+For standards and binding architecture, use [`doctoring.md`](doctoring.md), [`doctoring/browser-agent-protocols.md`](doctoring/browser-agent-protocols.md), [`PRD.md`](PRD.md), [`TRD.md`](TRD.md), [`product-roadmap.md`](product-roadmap.md), and linked ADR/UML/ERD/traceability records. Issues #199-#203 contain their own APA 7th standards and research traceability. This baseline intentionally records delivery state and never promotes planned adapters or active pull-request code to implemented behavior.
diff --git a/docs/product-technical-gap-baseline.md b/docs/product-technical-gap-baseline.md
index c4a87e6c7..8f478dc43 100644
--- a/docs/product-technical-gap-baseline.md
+++ b/docs/product-technical-gap-baseline.md
@@ -1,1297 +1,73 @@
 # Product and Technical Gap Baseline
 
-This is a dated delivery baseline, not a substitute for the PRD, TRD, roadmap, architecture decisions, or live GitHub state. It keeps buyer-visible gaps, current issues, active pull-request evidence, and commercial completion tracks in one discoverable place. Protected `main` is the implementation boundary: code in an open pull request is not shipped behavior.
+This file is the buyer-facing decision surface for the live OriginWeave delivery gap. Protected-main code, exact GitHub state, executable browser evidence, and immutable owner contracts remain authoritative; this document does not turn an active PR or a protocol command acknowledgement into shipped behavior.
 
-## Current live delivery state
+The detailed historical dossier that previously occupied this path is preserved byte-for-byte at [`docs/evidence/product-technical-gap-baseline-through-2026-09-09.md`](evidence/product-technical-gap-baseline-through-2026-09-09.md). Historical measurements are evidence for their recorded generation, not transferable acceptance for a later head.
 
-This volatile section is refreshed from live GitHub state and is authoritative only for the exact observations recorded here. The dated snapshot below remains historical evidence and is not promoted to current acceptance evidence. Live GitHub PR/base/head/check APIs are authoritative over PR bodies and prior maintenance prose; a body that still names an older head is stale evidence, not merge evidence.
+## Observed delivery cut — 2026-09-09 12:57 UTC
 
-### Latest verified cut: 2026-09-09
+This is a dated observation receipt, not a continuously live counter. The cut below includes this #309 Draft successor after its Ready-to-Draft repair. Live GitHub state supersedes this cut after 2026-09-09 12:57 UTC.
 
-#### Presentation identity and controlled-browser evidence
+- Protected-main truth remains signed-valid `87c4daa1830bac5a5228b6036752ad5633232085`.
+- The observed queue contains **130 open pull requests: 12 Ready/non-draft and 118 Draft; 14 open non-PR issues**. The three GitHub searches used for total/Ready/Draft counts and the issue search all returned `incomplete_results=false`.
+- Open issues: the 14 non-PR items above remain separate from the 130-PR delivery queue and are not counted as shipped product behavior.
+- Ruleset `18156473` is active. It requires one approving review, dismissal of stale reviews after push, review-thread resolution, an extra approval for unattributed changes, seven central required workflows, and blocks deletion/non-fast-forward updates.
+- GitHub Release inventory is empty. There is no immutable OriginWeave release to treat as commercial acceptance; tag/release state must still be re-read before any release claim.
+- The baseline owner branch was exact `16098a0eecc57bfc545d22ce632d95d1ffccfe67` when this navigation repair started. That generation preserved the former 186,500-byte baseline as implementation evidence but still placed the buyer gap matrix behind extensive dated history.
 
-At `2026-09-09T08:49:58Z`, the live inventory was **131 open pull requests:
-14 Ready/non-draft and 117 Draft; 14 open non-PR issues**. Protected `main`
-remains `87c4daa1830bac5a5228b6036752ad5633232085`; no GitHub Release exists.
-Ruleset `18156473` still requires one counted approval and seven central required
-workflows. Queued reviewer evidence is non-passing: discard queued, skipped,
-cancelled, absent, predecessor, synthetic, status-only, and model-only evidence.
+Phase 1 is **in progress**, not shipped. A successful source test, an active-PR head, a protocol ACK, or a clean security scan is not protected-main/browser/release acceptance by itself.
 
-Issue #292 remains the buyer-visible presentation-identity gap. Draft #293 is
-at exact head `6855e2578ae94279cc9ab4a14527b016e8c049ee` on Ready #229
-`024f63690cf05cfe6f0d4a430f0e18ea8fd2c4d6`. Its reusable lifecycle deliberately
-does not apply reduced motion without a complete restore or disposable-context
-owner. Thus command planning cannot be presented as full profile application.
-Draft successor #298 is at exact head
-`d01f45c2c8ac7b0fc4dbc3d3ada60238732cdf8c` on that #293 head. Its non-force
-merge of #305 repaired an older source contract so it now asserts both that
-ReducedMotion remains a discoverable protocol capability and that the reusable
-planner exposes no media-mutation command. This removes an API/documentation
-contradiction only; #298's Draft checks are skipped and it does not establish
-live Chromium transport, page observation, cleanup, protected-main shipment,
-or release acceptance.
-An active controlled evidence harness is implemented, but it does not establish a
-product Browser Session adapter: its exact browser lane failed 0/3 at session
-creation before navigation, so application, page-observed post-conditions, cleanup
-acceptance, and release remain unproven. Issue #212 owns the sandbox-helper contract;
-the product branch must consume it rather than copy or weaken workflow behavior.
+## Buyer gap matrix
 
-#### Transport-closure verification and test-integrity repair
-
-The current queue also leaves Ready #287 at
-`3975daf48e01a5e9d1cf9fb104a3be1aa03b0402` with exact-head native and central
-checks queued. Its prior CodeQL dispatch outcome was an absent central verdict,
-not a classifier-source finding. No release, protected-main shipment, or approval
-is inferred from local work, a child merge, or queued checks.
-
-### Previous verified cut: 2026-09-08
-
-#### Published intent acknowledgment verification
-
-Published #271 head
-[`46db0045904f0289738df843d0a2f179c26673d3`](https://github.com/ContextualWisdomLab/OriginWeave/commit/46db0045904f0289738df843d0a2f179c26673d3)
-retains parent #270 `8eda96915dbbe4cc617f834267c7464689c2844d` and is **not shipped**.
-The previous head's formatter and coverage failures were reproduced. The repair
-retains connection and command pre-consumption guards, removes only a proven
-unreachable duplicate check, and tests genuine socket replies, pending-request
-preservation and private-text-safe diagnostics.
-
-Frozen-source local coverage is **1406/14894/19018/1552**
-(functions/lines/regions/branches), all 100%. Nine focused integration tests,
-148 repository contracts, complete workspace coverage tests, strict Clippy,
-formatting, five doctests and warnings-denied rustdoc pass. Actual Edge visual
-inspection verified the rendered acknowledgment contract and readable layout.
-Hosted [run 34186280263](https://github.com/ContextualWisdomLab/OriginWeave/actions/runs/34186280263)
-is terminal SUCCESS on this exact head: Rust contracts and Production coverage
-both pass. Both review threads are resolved and GitHub reports the Draft head
-**CLEAN**. Protected-parent adoption, governing approval and real browser
-outcome evidence remain required; this is still **not shipped**.
-
-Separately, #255 advanced from superseded test-only head
-`10d5e1ff78d46bbad004d4e0971d0ffceac757fa` to published repair
-`6b6c90ed3919ea84b69527ab688a087eeb45224d`. A real server-sent Close 1010
-first reproduced RED because the client mirrored the client-only code. The
-role guard now rejects it before reply or closure evidence while the paired
-server 1011 control remains valid. Exact-head local coverage is
-**1105/11574/14787/1218**, all 100%; 142 repository contracts, strict Clippy,
-formatting, doctests and warnings-denied rustdoc pass. Actual Edge inspection
-verified the rendered role, masking-key, control-budget and deadline contract.
-Hosted [run 34188785932](https://github.com/ContextualWisdomLab/OriginWeave/actions/runs/34188785932)
-is terminal SUCCESS: Rust contracts and Production coverage both pass on that
-exact head. The Draft is CLEAN, but this remains neither protected-main nor
-shipped behavior.
-
-Draft #293 advanced from visually inspected predecessor
-`6b5241c164f5283f8dd51b1846ef0e4dacec0b29` to exact head
-`0c077445d73640a6299ea4d379faa4b0ab0226c2` on parent #229. Its executed RED showed that standard BiDi was
-overstating complete Screen and Languages support. The minimal repair retains
-only Viewport, DevicePixelRatio, TimeZone and ReducedMotion, pins the immutable
-3 September 2026 dated W3C Working Draft identity, and adds an explicit typed
-viewport/DPR reset intent because session teardown does not clear that override.
-The current head binds three command intents for those four surfaces plus one
-cleanup intent to one bounded opaque browsing context without sending them or
-minting evidence. Exact local coverage is **594/4974/5980/754**, all 100%; 159
-repository contracts, strict Clippy, formatting, doctests and warnings-denied
-rustdoc pass. Exact-head Edge inspection was attempted, but the foreground Edge
-tab was a different repository and was not navigated; current-head visual acceptance is unproven.
-Generated-rustdoc inspection is separately blocked by local-file browser policy.
-The Draft is CLEAN; real Chromium application and page-observed evidence remain
-in #292.
-
-Both source writer leases are released. Inventory remains 126 open PRs, 14
-issues and zero releases. Earlier sections below retain dated evidence, not the
-current status of their superseded runs.
-
-#### Published repeated-control repair
-
-Published #255 head
-[`3a8e6f4f89db2f53c144adb3351c153d89adca58`](https://github.com/ContextualWisdomLab/OriginWeave/commit/3a8e6f4f89db2f53c144adb3351c153d89adca58)
-retains parent #252 `363a78e36e7690e9ed5bf49829567e00e2ec5d59` and is **not shipped**.
-The two-Pong real-peer case first failed. The observer now handles up to **64**
-pre-Close Ping/Pong frames under one total deadline, with fresh caller-supplied keys
-consumed only by Ping responses. A separate key remains reserved for Close.
-Tests prove exact-budget success, **65th** control rejection, exhausted and reused
-key rejection without an extra reply, and literal separately masked Pong payloads.
-
-Frozen-final-source local coverage is **1104/11553/14758/1216**
-(functions/lines/regions/branches), all 100%. All **20 closure** integration tests,
-six controlled-clock tests, 142 repository contracts, strict Clippy, doctests, and
-warnings-denied rustdoc pass. Actual Edge inspection verified the rendered key-slice
-API, local resource limit and total-deadline contract. Hosted
-[run 34184829970](https://github.com/ContextualWisdomLab/OriginWeave/actions/runs/34184829970)
-is queued at this checkpoint: local verification is **not hosted acceptance**.
-
-The repeated-control implementation gap is closed locally, not released. Current-head
-review/check acceptance and protected-parent adoption remain required; **process exit**
-and **profile cleanup** still lack runtime evidence. Inventory remains 126 open PRs,
-14 issues and zero releases. The following same-day sections preserve earlier
-checkpoints; their pending implementation claims are superseded by this section.
-
-#### Published deadline and Close-code repairs
-
-Published #255 head
-[`8716b9d441960a112446c1f89ab417ad9abe28d2`](https://github.com/ContextualWisdomLab/OriginWeave/commit/8716b9d441960a112446c1f89ab417ad9abe28d2)
-retains parent #252 `363a78e36e7690e9ed5bf49829567e00e2ec5d59` and is **not shipped**.
-One operation-wide monotonic deadline now covers the complete closure exchange.
-A delayed real-peer test first reproduced budget renewal; controlled-clock tests
-then verified every deadline transition and final evidence admission.
-
-The shared Close validator also rejects unassigned protocol codes **1016–2999**
-against the current IANA registry, while preserving application/private ranges.
-The boundary test first failed on 1016; real peers now verify no reply or closure
-evidence for 1016, 2000, and 2999. Local functions/lines/regions/branches are
-**1101/11492/14662/1218**, all 100%. All **18 closure** integration tests,
-**142 repository** contracts, strict Clippy, doctests and warnings-denied rustdoc pass.
-Hosted [run 34183766437](https://github.com/ContextualWisdomLab/OriginWeave/actions/runs/34183766437)
-is pending at this checkpoint: local verification is **not hosted acceptance**.
-
-The fresh inventory remains 126 open PRs (12 Ready, all BLOCKED; 114 Draft),
-14 issues and zero releases; protected main remains
-`87c4daa1830bac5a5228b6036752ad5633232085`. The current rule still requires one
-counted approval and seven central workflows. Remaining #255 work is **repeated Ping/Pong**
-with fresh masking keys; browser **process exit** and **profile cleanup** remain unproven.
-Publication still requires protected source, an explicit version and complete release
-artifacts/evidence. The earlier checkpoint below records its own earlier state,
-not the current implementation gaps.
-
-#### Transport-closure verification and test-integrity repair
-
-The latest inventory contains **126 open pull requests: 12 Ready/non-draft and
-114 Draft; 14 open non-PR issues**. All 12 Ready candidates remain BLOCKED.
-Protected main remains `87c4daa1830bac5a5228b6036752ad5633232085`, with
-no package release. These counts do not establish review-thread resolution.
-
-PR #255 has adopted parent #252 at
-`363a78e36e7690e9ed5bf49829567e00e2ec5d59`; the previous instruction to perform
-that adoption is now historical. It remains Draft and is not protected-main acceptance.
-
-- Verified predecessor: [`d126242c7198c447d0fab7983d529441340fd1c9`](https://github.com/ContextualWisdomLab/OriginWeave/commit/d126242c7198c447d0fab7983d529441340fd1c9)
-  passed both Rust contracts and Production coverage in
-  [run 34179452950](https://github.com/ContextualWisdomLab/OriginWeave/actions/runs/34179452950).
-  Full local functions/lines/regions/branches were **1090/11218/14328/1210**, all 100%.
-- Published follow-up: [`07ef43ec71b6dbd8540629bf1df5a63b81541ee4`](https://github.com/ContextualWisdomLab/OriginWeave/commit/07ef43ec71b6dbd8540629bf1df5a63b81541ee4)
-  corrects three masking-rejection test peers. Propagating peer failures first
-  reproduced two false-green fixtures; all 16 closure tests and the full workspace
-  suite including doctests then passed. Hosted
-  [run 34180304951](https://github.com/ContextualWisdomLab/OriginWeave/actions/runs/34180304951)
-  was queued at this checkpoint; predecessor coverage does not prove this head passed.
-
-The direct writer tests compare literal Close bytes with and without a status code,
-invalid deadlines, and adjacent masking-key rejection. The repaired observer fixtures
-consume the real preceding text/Pong before testing reuse and verify no rejected
-response bytes. Broad transport errors with ignored server failures were insufficient.
-
-Remaining buyer-visible gap: repeated Ping/Pong traffic and an operation-wide deadline
-still need implementation and acceptance; unassigned Close status handling needs
-protocol review. Browser-process exit and profile cleanup remain unproven. The next
-source work belongs to #255 before downstream adoption; no PR is closed as a substitute
-for carrying its delta. Actual Edge inspection verified the PR page and corrected
-evidence presentation, not the OriginWeave product browser. Package publication remains
-withheld until protected-source, version, artifact and release requirements are met.
-
-### Historical verified cut: 2026-09-07
-
-#### Published session-end reply binding: 05:35 UTC
-
-The refreshed five-page inventory contains **125 open pull requests: 12 Ready/non-draft
-and 113 Draft; 14 open non-PR issues**, with zero unresolved review threads and no
-pagination gaps. All Ready candidates remain BLOCKED. Protected main remains
-`87c4daa1830bac5a5228b6036752ad5633232085`; no tags or releases exist. Active ruleset
-`18156473` requires one counted approval and seven central workflows; the author
-remains the only listed collaborator. This reviewer-provisioning gap permits no bypass.
-
-Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
-
-| Owner | Publication | Exact head | Parent | Local coverage (F/L/R/B) |
-| --- | --- | --- | --- | --- |
-| #251 | Published; Draft | [`924ad975`](https://github.com/ContextualWisdomLab/OriginWeave/commit/924ad97551750d4a901ded38b89488cc5438e54f) | #250 `bbdc6ace` | 1057/10840/13873/1194 |
-| #252 | Published; Draft | [`363a78e3`](https://github.com/ContextualWisdomLab/OriginWeave/commit/363a78e36e7690e9ed5bf49829567e00e2ec5d59) | #251 `924ad975` | 1064/10898/13947/1194 |
-
-Both heads passed all four coverage dimensions at 100%, 142 Python contracts,
-compileall and full stable Rust formatting, check, tests, strict Clippy and
-dependency-inclusive warnings-denied rustdoc. Actual Edge inspection covered the
-rendered APIs, the end-response link to its sealed-reply type, and published commits.
-Independent source review found no actionable findings; it is not counted approval.
-
-#251 preserves its sender and original tests while adopting status-reply safeguards.
-#252 reproduced replacement end-reply acceptance before and after parent adoption,
-then reused connection-bound registration and sealed reply correlation. Replacement
-success and error replies now leave the original request pending, and its original
-reply can still complete it. The fixtures use retained receipts after separate
-listener servers finish: they do not establish same-endpoint or live-stream recovery,
-browser shutdown, profile cleanup or protected delivery.
-
-Coverage artifact SHA-256 for #251 is
-`5cbb52b50b5740a22d269490f6296380d4b53bbe61ceae632376f6de63ff1054`;
-for #252 it is `9f2249637f31916acf9874baec724ddedfe621e5b14fdb6f4bd850620e89f307`.
-Exact-head CI `34085877650` (Rust `101629699915`, coverage `101629699750`) and
-`34087239755` (Rust `101633511589`, coverage `101633511320`) are queued.
-Local verification is not hosted acceptance. Both source writer leases are released.
-
-Next, safe successor #255 must adopt #252 while preserving its stronger closure
-provenance; #253/#254 remain open and are not safe intermediate prerequisites.
-New issue #292 owns version-pinned browser presentation evidence after #229; its
-requested implementation and real-browser acceptance remain unverified. Existing
-release, runtime, extraction, API and enterprise gaps remain open. Automatic package
-publishing is still unattempted: the protected package disables publication, no
-Python package or registry-publishing workflow exists, and release evidence plus an
-explicit version decision are required. Secret metadata is not release readiness;
-organization key availability remains unverified. No gates or credentials changed.
-Earlier checkpoints remain unchanged historical evidence.
-
-#### Published status-response repair: 04:45 UTC
-
-The refreshed five-page inventory contains **125 open pull requests: 12 Ready/non-draft
-and 113 Draft; 13 open non-PR issues**, with zero unresolved review threads and no
-pagination gaps. All Ready candidates remain BLOCKED. Protected main remains
-`87c4daa1830bac5a5228b6036752ad5633232085`; no tags or releases exist. Active ruleset
-`18156473` requires one counted approval and seven central workflows; the author is
-the only listed collaborator. This reviewer-provisioning gap does not permit bypass.
-
-Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
-
-| Owner | Publication | Exact head | Parent | Local coverage (F/L/R/B) |
-| --- | --- | --- | --- | --- |
-| #250 | Published; Draft | [`bbdc6ace`](https://github.com/ContextualWisdomLab/OriginWeave/commit/bbdc6ace7a5932adf24836700f806850e6b230bc) | #249 `65ac3ab9` | 1043/10711/13717/1188 |
-
-All four coverage dimensions are 100% locally. Prior exact-head CI `34064507212`
-failed formatting and coverage; that terminal evidence supersedes its historical
-queued description. The repair reuses the sealed-reader tests and a shared, pure
-reply-routing check. Validation and connection provenance still precede completion.
-Public loopback checks reject events, unattributable errors and unbound requests
-without consuming pending work. After an unbound request rejects a sealed reply,
-the original request can still accept it. This does not prove a browser outcome.
-
-All 142 Python contracts, compileall and full Rust formatting, check, tests, Clippy
-and warnings-denied workspace documentation passed at the published head. Coverage
-artifact SHA-256: `e1fccefc6b56eabe653ae41377fbcec0aa89b4be1ada92dba73d3ef87e841029`.
-Actual Edge inspection confirmed readable API documentation and the published commit.
-Independent source review found no actionable findings; it is not counted approval.
-
-Exact-head CI `34084134654` has Rust job `101624834145` and coverage job `101624834026`
-queued. Local verification is not hosted acceptance. The source writer lease is
-released; this documentation checkpoint needs its own verification and visual inspection.
-Earlier checkpoints remain unchanged historical records, not current acceptance.
-
-Automatic publishing remains unattempted: the protected package disables publication,
-no Python package or publishing workflow is present, and release evidence and an
-explicit version decision are still required. Empty repository/environment secret
-metadata does not prove organization keys absent; their listing remains unavailable
-under current permissions. No credential values, workflow gates or release settings
-were changed. Next executable work remains parent-first status-receipt adoption,
-protected foundations and release-readiness evidence under their separate owners.
-
-#### Published semantic-action adoption: 04:11 UTC
-
-The five-page inventory contains **125 open pull requests: 12 Ready/non-draft
-and 113 Draft; 13 open non-PR issues**, with zero unresolved review threads and no
-pagination gaps. All Ready candidates remain BLOCKED. Protected main remains
-`87c4daa1830bac5a5228b6036752ad5633232085`; no tags or releases exist.
-Active ruleset `18156473` still requires one counted approval and seven central
-workflows; the only listed collaborator is the author. Advisory review is not approval.
-
-Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
-
-| Owner | Publication | Exact head | Parent | Local coverage (F/L/R/B) |
-| --- | --- | --- | --- | --- |
-| #93 | Published; Draft | [`82056d13`](https://github.com/ContextualWisdomLab/OriginWeave/commit/82056d13aa94c106060b84ee76be56fcb7787fc8) | #271 `b0410ae9` | 1403/14852/18976/1552 |
-| #95 | Published; Draft | [`6b29d890`](https://github.com/ContextualWisdomLab/OriginWeave/commit/6b29d890245ed2f612c2998198f4e8c8a06da312) | #93 `82056d13` | 1408/14897/19022/1552 |
-| #96 | Published; Draft | [`cbabf55c`](https://github.com/ContextualWisdomLab/OriginWeave/commit/cbabf55c6a25b979fa0d9e3c1677338665975ab7) | #95 `6b29d890` | 1410/14908/19030/1552 |
-
-Each listed coverage dimension is 100% locally. These ordinary parent adoptions retain
-node-action binding, explicit policy allow and dispatch-time node validation while
-inheriting connection-bound reply rejection. The original connection can still complete
-its own pending request; unrelated pending work survives. The child implementations
-and original tests were preserved. Node authority, policy decisions, callback execution
-and reply provenance remain separate checks, not proof of the requested browser outcome.
-
-#93 passed 10 focused Rust tests and 149 Python contracts; #95 passed eight focused tests
-and 150 Python contracts; #96 passed six focused tests and 151 Python contracts. Complete
-local Rust gates passed. #93's final Python/Markdown-only correction preserved the entire
-measured Rust tree and manifests. Coverage artifact SHA-256 values, in owner order:
-`9479014b94c1575a379d31fea09309deaf00143374b6dd6f04a539aaf452a685`,
-`666797dec8486f0f196616525303bd24dc52ef5d035f1c548e1bddde3fb48a22`,
-`f69829e1968a75c8d3c8ea7d08df87669d47a7d6adfbbb88d08d9bb0d384b10d`.
-
-Exact-head CI `34079739018`, `34080772063` and `34081979602` remain queued.
-#93's MV3 run `34079739021` is also queued. Independent read-only review found no
-remaining actionable findings; it is not counted approval. Source leases are released.
-All three revisions received actual visual inspection in isolated Edge, with readable
-API documentation and no observed clipping or overlap. Dependency-inclusive workspace
-rustdoc restored cross-crate links omitted by the no-dependencies build. This is
-documentation presentation, not product-browser acceptance.
-
-The user-requested `AGENTS.md` lessons were published at `de06d65e` with 197 Python
-contracts and their own rendered visual inspection. Automatic package publishing was
-checked but not attempted: protected main's core manifest has `publish = false`, no
-publishing workflow or Python package manifest is present, and the release contract
-still requires its evidence and an explicit version decision. Repository/environment
-secret metadata lists were empty; organization-secret listing returned 403, so those
-keys remain unverified, not proven absent. No secret values were accessed.
-
-Next are status receipts, protected-foundation work and release-readiness evidence.
-#195/#279 workflow recovery and #212 sandbox acceptance retain their separate owners.
-This baseline requires its own exact-head verification and visual inspection; it cannot
-inherit source coverage or hosted acceptance. Earlier checkpoints below are unchanged
-historical observations. No merge, package publication or release is claimed.
-
-#### Published observation safeguards: 03:00 UTC
-
-The complete five-page inventory contains **125 open pull requests: 12 Ready/non-draft
-and 113 Draft; 13 open non-PR issues**, with zero unresolved review threads and no
-pagination gaps. All Ready candidates remain BLOCKED. Protected main remains
-`87c4daa1830bac5a5228b6036752ad5633232085`; no tags or releases exist.
-Active ruleset `18156473` requires one counted approval and seven central workflows;
-the sole collaborator is the author. Advisory review does not satisfy that approval.
-
-Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
-
-| Owner | Publication | Exact head | Parent | Local coverage (F/L/R/B) |
-| --- | --- | --- | --- | --- |
-| #270 | Published; Draft | [`8eda9691`](https://github.com/ContextualWisdomLab/OriginWeave/commit/8eda96915dbbe4cc617f834267c7464689c2844d) | #269 `3df2a631` | 1346/14113/17928/1464 |
-| #271 | Published; Draft | [`b0410ae9`](https://github.com/ContextualWisdomLab/OriginWeave/commit/b0410ae92bd20eaf31d09b7d49390e13cb045999) | #270 `8eda9691` | 1393/14770/18899/1546 |
-
-Every listed coverage dimension is 100% locally. #270 repaired wrong-session dispatch,
-invalid deadlines and reused masking keys through existing transport guards, preserving
-the fixed observation command and ordinary parent history. Its 11 focused tests,
-147 Python contracts and complete stable Rust gates passed. Coverage artifact SHA-256:
-`7748d42c541140314995ffa2087b3aca87d52875f98a6ffa36614a0453b30260`.
-
-#271 reproduced replacement-connection completion at RED `29cd0d66`, still failing after
-parent adoption `14efb678`. The existing sealed receipt and connection-aware correlation
-now reject foreign success, protocol error and script exception. A same-listener test
-proves the original connection can still complete its command, leaving unrelated work
-pending. Earlier fixture compile errors are not runtime RED evidence. All 148 Python
-contracts, focused recovery tests and complete stable Rust gates passed. Coverage SHA-256:
-`8f211170f9e7b691b95b12151e9e87827e4cb741ca8fa2892576adf8421e8a30`.
-
-Exact-head native CI `34076117534` and `34077987302` remain queued, not passing evidence.
-Independent read-only review found no remaining actionable findings; it is not counted
-approval. Source leases `5563898819` and `5564211278` are released. Both revisions received
-actual visual inspection in isolated Edge: generated API documentation and published
-commit views were readable without observed clipping or overlap. This is documentation
-presentation, not product-browser acceptance. This baseline needs its own verification
-and visual inspection; it cannot inherit source coverage or hosted acceptance.
-
-Next are downstream semantic-action adoption, status receipts and protected-foundation
-work. Matching text replies do not establish authenticated browser ownership, policy
-approval or complete authorized action success. #195/#279 workflow recovery and #212
-sandbox acceptance retain their separate owners. Earlier checkpoints below are unchanged
-historical observations, including superseded next actions; no release is claimed.
-
-#### Published response and observation: 01:45 UTC
-
-The fresh five-page inventory still contains **125 open pull requests: 12 Ready/non-draft
-and 113 Draft; 13 open non-PR issues**, with zero unresolved review threads.
-All Ready candidates remain BLOCKED.
-Main remains `87c4daa1830bac5a5228b6036752ad5633232085`; there are no tags or releases.
-Active ruleset `18156473` requires one counted approval and seven central workflows;
-the sole collaborator is the author. No advisory review is counted as that approval.
-
-Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
-
-| Owner | Publication | Exact head | Parent | Local coverage (F/L/R/B) |
-| --- | --- | --- | --- | --- |
-| #268 | Published; Draft | [`ff27220c`](https://github.com/ContextualWisdomLab/OriginWeave/commit/ff27220cb5eb4d11ca1dc5614a4181e1a397a3f1) | #267 `ebd507ae` | 1325/13907/17682/1456 |
-| #269 | Published; Draft | [`3df2a631`](https://github.com/ContextualWisdomLab/OriginWeave/commit/3df2a631bacd7109b3982fdd7ac599d0bd92a589) | #268 `ff27220c` | 1338/14025/17843/1460 |
-
-Each dimension is 100% covered locally. #268's actual RED `716fd842` reproduced three
-pointer/session failures before ordinary parent adoption. The sealed text consumer and
-all original text tests were preserved. Review exposed an unqualified historical limitation;
-RED `49d18f5f` reproduced it, and the correction anchors it to predecessor `35cb1197`.
-#269's RED `658fb676` reproduced dispatch to the wrong browser session before ordinary
-merge `f883fd6f`. Its fixed field-observation command and all eight child tests remain
-byte-identical to the predecessor; the entire network crate matches its parent.
-
-#268 passed 20 focused tests and 146 Python contracts; #269 passed 18 focused tests and
-147 Python contracts. Both passed complete local Rust 1.97.1 gates, compileall, CodeGraph
-and diff checks. Coverage artifact SHA-256 values are respectively
-`bce553c93d5b5dd3cb59c78bbb75c9f1225b659ec7ea22b674c54455e5067a07` and
-`1e203e6c0801bedbbcdbeb1be500b533a3d68e0ac842de221f7f470c5e742c66`.
-The pinned nightly branch-instrumentation warning remains visible.
-
-Exact-head native CI `34072645796` and `34073733364`, and #269 MV3 `34073733355`,
-are queued, not passing evidence. Independent read-only reviews found no remaining
-actionable findings. Source leases `5563501468` and `5563726066` are released.
-Both published PRs and dated dossiers received actual visual inspection in isolated Edge:
-revision links, current/historical separation and text were readable without visible
-clipping or overlap. This is GitHub documentation presentation, not product-browser acceptance.
-This baseline requires its own exact-head tests and visual inspection.
-
-Next is #270's content-aware adoption of #269, followed by response/value verification,
-status receipts and protected-foundation work. Request construction and protocol replies
-still do not prove field changes, authenticated browser ownership, policy approval or
-authorized action success. #195/#279 workflow recovery, #212 sandbox acceptance and
-.github#1929 dispatch identity admission retain their separate owners. Earlier checkpoints
-below remain unchanged historical observations, including their superseded next actions.
-
-#### Published input descendants: 00:40 UTC
-
-The latest complete five-page review/check/thread inventory contains **125 open pull requests:
-12 Ready/non-draft and 113 Draft; 13 open non-PR issues**, with zero unresolved review threads
-and no unfinished pagination. All 12 Ready candidates remain BLOCKED. Current rules require
-one counted approval and seven required workflows; the sole collaborator is the author.
-Protected main remains `87c4daa1830bac5a5228b6036752ad5633232085`, with no tags or releases.
-REST release inventory was rate-limited; GraphQL independently confirmed both empty inventories.
-
-Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
-
-| Owner | Publication | Exact head | Parent | Local coverage (F/L/R/B) |
-| --- | --- | --- | --- | --- |
-| #265 | Published; Draft | [`e94a2372`](https://github.com/ContextualWisdomLab/OriginWeave/commit/e94a2372fe3771f9ddf70291d34fc9a7e4770ec9) | #264 `43395711` | 1297/13618/17315/1442 |
-| #266 | Published; Draft | [`e3885f69`](https://github.com/ContextualWisdomLab/OriginWeave/commit/e3885f69df2cf3899184209efdee5b11bba1bd86) | #265 `e94a2372` | 1310/13766/17524/1452 |
-| #267 | Published; Draft | [`ebd507ae`](https://github.com/ContextualWisdomLab/OriginWeave/commit/ebd507ae56c3064e3cae5566502f539c20618a8f) | #266 `e3885f69` | 1318/13852/17610/1456 |
-
-Functions, lines, regions and branches are each 100% covered on these exact local source trees.
-#265 replay `e7fb1527` first failed both replacement-reply tests before ordinary merge
-`d847b530`. The contributor's existing outbound-session guard remains intact. #266 replay
-`009f9a41` and #267 replay `4e020e16` each failed all three inherited pointer regressions
-before ordinary merges `d9503b30` and `7e4bd76d`. The repaired click path rejects a foreign
-session before reserving or sending work, and rejects replacement success/error replies
-without consuming either pending request. The original reply completes only its own request.
-
-#265 passed 14 focused tests; #266 passed 21; #267 passed 23 focused tests. Each exact head
-passed full local Rust 1.97.1 gates, 145 Python contracts, compileall, CodeGraph and diff checks.
-The text constructor, privacy checks and eight core text tests were preserved in #266.
-#267 also preserves its existing text sender, public exports and ten text-send tests unchanged.
-Coverage artifact SHA-256 values for #265, #266 and #267, respectively, are
-`fc9aebce5e633be1c4faa10755e56bb792a0f2679daeef3a0a1984d4a4b58404`,
-`a06aebb7c2f30c153c3a3867f3b7847d9c485ae2db5ba2b23029ca0375e4ec79` and
-`8bd32ad83ba5f9ef65bad8a3503f5e6cff38335b8526bff9f974bd6ee7fabceb`.
-The pinned nightly branch-instrumentation warning remains explicit; no coverage gate changed.
-
-Native CI `34068277243`, `34068882527` and `34070121583` remains queued for those three heads;
-#265/#266 MV3 runs `34068277244` and `34068882439` are also queued. Independent read-only
-preservation reviews found no actionable findings, but are not counted approval. Source writer
-leases `5563064004`, `5563186050` and `5563277731` are released. Actual visual inspection
-used actual screenshots of published exact-head evidence: in-app for #265/#266 and isolated
-installed Edge for #267 after browser-control connection failures. Revisions, caveats and
-wrapping were readable without observed clipping or overlap. This is GitHub presentation,
-not product-browser acceptance. This baseline needs its own checks and visual inspection.
-
-Next source work is #268's content-aware adoption of #267, retaining its existing sealed
-text-response consumer rather than rebuilding it. Then continue status receipt prerequisites
-and protected-main asset preservation at their canonical owners. #195/#279 workflow recovery,
-#212 sandboxed browser acceptance and .github#1929 dispatch identity admission remain separate.
-Text transport and pointer acknowledgments do not prove field values, browser authentication,
-policy approval, navigation causality or release acceptance. Earlier dated observations below
-remain unchanged historical evidence and do not override this checkpoint.
-
-### Latest verified cut: 2026-09-06
-
-#### Published subscription descendants: 23:27 UTC
-
-The complete five-page live inventory contains **125 open pull requests: 12 Ready/non-draft and
-113 Draft; 13 open non-PR issues**. It has zero unresolved review threads and no
-unfinished review/check/thread pagination. This supersedes the older sole-#147
-thread observation below. All 12 Ready candidates remain BLOCKED. The active ruleset
-still requires one counted approval and seven required workflows; only one repository
-collaborator is available, so independent reviewer provisioning remains unresolved.
-Protected main remains `87c4daa1830bac5a5228b6036752ad5633232085`; there are no tags or releases.
-
-| Owner | Publication | Exact head | Parent | Local coverage (F/L/R/B) |
-| --- | --- | --- | --- | --- |
-| #263 | Published; Draft | [`4868d3e9`](https://github.com/ContextualWisdomLab/OriginWeave/commit/4868d3e9f19133ac3382ee8532878aef27468893) | #277 `46ae62aa` | 1244/13053/16729/1422 |
-| #264 | Published; Draft | [`43395711`](https://github.com/ContextualWisdomLab/OriginWeave/commit/433957117ad9e29b26715b062f5adcc9789744ba) | #263 `4868d3e9` | 1282/13461/17140/1440 |
-
-F/L/R/B denotes functions, lines, regions and branches; every metric is 100% covered.
-
-Both owners adopted their parents by ordinary merges after actual replacement-reply
-regressions failed. #263 replay `90395f81` failed 0/2, followed by merge `9e85cadc`
-and sealed-reader fixture adaptation `cb0c4261`. Its 10 focused tests and full local
-Rust 1.97.1 gates passed, alongside 144 Python contracts. #264 replay `637fd97d`
-failed 0/2 before merge `92e576c8`; its 13 focused tests, full local Rust 1.97.1 gates
-and 145 Python contracts passed. Both passed compileall, CodeGraph and diff checks.
-The unchanged pinned coverage verifier passed the counts above; experimental branch
-instrumentation remains explicit. Their coverage artifacts have SHA-256 values
-`1e51cd7d08ebacc80ecadb568e0bca94384ffc0e1d721237b018cae31daa1336` and
-`e102bda45da886bc3138b981dac73f4239e2f6e5cd893688f10ed72b1f5af771`, respectively.
-
-The child keeps its original browser-state association, increasing request identifiers,
-separate raw/typed traffic and consuming subscription-shutdown ownership. Original
-replies still complete their own request after a foreign reply is rejected; unrelated
-work remains pending. Independent read-only preservation reviews found no actionable
-findings, but are not counted GitHub approval. Writers `5562611932` and `5562770321`
-are released. CI `34065055213` for #263 and CI `34066516991` plus MV3 `34066516992`
-for #264 were queued at the recorded readbacks; predecessor success is not new-head proof.
-
-Actual visual inspection used actual in-app screenshots of both published exact-head
-traceability sections: wrapping and historical/current boundaries were readable without
-observed clipping or overlap. This is GitHub presentation evidence, not product-browser acceptance.
-Earlier locked-Mac and pending-visual observations remain historical. This new baseline
-revision needs its own checks and visual inspection; source coverage does not validate it.
-
-Central dispatch identity admission remains separately owned by .github#1929
-([verified diagnostics](https://github.com/ContextualWisdomLab/.github/issues/1929#issuecomment-5562597667));
-the documented actor/sender mismatch is not evidence of recovery. Do not change owner
-settings or repeat a disproven dispatch route under this documentation task. Next safe
-source work is content-aware descendant adoption, pointer outbound authority and status
-receipt prerequisites, alongside protected-main asset preservation. Browser authentication,
-navigation-event authenticity, causal page effects and release acceptance remain unproven.
-
-#### Connection-bound text responses: 14:54 UTC
-
-This checkpoint supersedes the source status and executable actions in the earlier
-timed checkpoints below, which remain historical evidence. The fresh complete
-five-page inventory still contains **125 open pull requests: 12 Ready/non-draft and
-113 Draft; 13 open non-PR issues**. All Ready candidates remain BLOCKED; only #147 retains unresolved
-thread `PRRT_kwDOTulPlM6coZwc`. Protected main is unchanged at
-`87c4daa1830bac5a5228b6036752ad5633232085`; release and tag inventories are empty.
-
-Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
-
-Published sender #267 `3346d8ecc72932b98ec495d9cc52d6e5727c3064` remains on #266
-`eb6c236ff2f4a58b807a2f2c914bd1ddb6079fb3`. Published response owner
-#268 `e567af9e678fd4791776df795e89ed666975e6c2` adopts it by ordinary merge
-`e1188c86`. Actual socket RED `4632f2df` accepted a reply from a replacement
-connection using the same listener/session. Sealed consumer `d6889c80` exposed
-missing sender provenance; the parent now records its private connection generation
-before I/O and the consumer checks the sealed receipt before consuming pending work.
-Foreign success/error replies retain both pending entries; the original reply still
-completes its request and leaves unrelated work pending. Seven focused response tests
-preserve real remote-error consumption, extensible success, malformed/unknown
-retention, family isolation and missing-provenance rejection.
-
-Both exact source heads pass full local Rust1.97.1 gates, 145 Python contracts,
-compileall, CodeGraph and diff checks. Their function/line/region/branch coverage is
-1317/13835/17598/1456 for #267 and 1324/13890/17670/1456 for #268, each 100%.
-Coverage SHA-256 values are respectively
-`8d7a3687557d067ee99fd512936077c0c9662d40d9d6544232bb5a377dc88e23` and
-`5865d6e69dc5584bad8ef4e866970eb6aa845ad692aa5421a0ffb6f6f01e1540`.
-CI `34040202356` and `34040306372` remain queued; formal reviews are empty.
-Actual Edge screenshots verified the published source PRs' exact heads, Draft state
-and readable evidence without observed clipping or overlap. This is GitHub
-presentation evidence, not product-browser acceptance or protected-main delivery.
-Writers `5559965531` and `5559948553` are released. This documentation revision
-requires its own publication, visual inspection and exact-head checks.
-
-At the 14:54 checkpoint, next source work was to address pointer and status receipt provenance in their existing
-owners, plus the separate pointer outbound session-binding gap, before promoting
-their evidence. Reuse the sealed reader and canonical connection checks; status
-projection must still validate before correlation consumption. Browser ownership,
-policy approval and observed field/DOM postconditions remain unfinished. Current
-foundation #195 `63997bcf555e2c5c8e91ba287734ffba3837a1b7` remains on
-`6922dd98779e8f8aad132a3b1f563d7ba6e6d070`; #279 workflow generation and #212
-sandbox boundaries remain prerequisites. Counted approval and central workflow
-requirements must not be bypassed. Source and documentation writers stay separate.
-
-##### Pointer receipt follow-up: 15:28 UTC
-
-This later checkpoint supersedes the preceding pointer next-action claim, not the
-dated text-entry evidence. Published sender #257 `9451fd8a23dec95b31749376bc78c2eaca977fe8`
-remains based on #256 `881c7f09ee9161ce8664dd75226938ecf60b85e5`; published
-response owner #258 `5417ce32ed957aa166807f1023647caccc2920cb` adopts that sender
-by ordinary merge `0234b587`, preserving the existing hourly writer's test history.
-
-Executed RED `8193fcd5` accepted a replacement connection's reply. Stronger RED
-`d9396f05` held the listener address and session constant and reproduced both foreign
-success and error consumption. Consumer-only `588fe731` failed original-response
-acceptance with missing sender provenance. The integrated sender registers its
-private connection generation before I/O; the consumer validates the existing sealed
-receipt before consuming pending work. Six focused tests pass, including exact
-foreign-connection mismatch, unchanged pending work and original-reply recovery
-with counts **2 → 2 → 1**, extensible success, real remote error, malformed input
-and unknown-id retention. No public receipt constructor or raw fallback was added.
-
-Both source heads passed full local Rust 1.97.1 gates, 142 Python contracts,
-compileall, CodeGraph and diff checks. Exact function/line/region/branch coverage is
-1093/11134/14204/1214 for #257 and 1100/11191/14276/1214 for #258, each 100%.
-Their artifact SHA-256 values are
-`a976c273d86d43eef366044cca5b73566064844612a35ba02ac4b7fe59a0969f` and
-`ed46b5881d952723eec724a74e6ebb8856a0801e4db0570c9f8a6e6cf04eadb5`.
-CI `34041977863` and `34042223733` remain queued; successful advisory statuses
-are not counted approval. Actual Edge screenshots verified both published PR
-heads, parent boundaries and readable evidence without observed clipping/overlap.
-This is GitHub presentation evidence, not product-browser acceptance. Writers
-`5560144371` and `5560167022` are released. This baseline update requires its own
-publication, rendered inspection and hosted checks; source coverage is not its proof.
-
-The next source queue is content-aware descendant adoption, separate pointer
-outbound session authority, and status receipt provenance. Existing status owners
-#249 `84b9407978ae0f6c115f01170b6069c601b21104` and
-#250 `7e85a7e5f0147f4b712129cd19aaa3d0a0a54634` precede the shared sealed-receipt
-and connection-registration helpers: establish the canonical prerequisites in
-dependency order rather than copying a later descendant or creating a cyclic stack.
-Status projection must remain validated before state consumption. Browser ownership,
-policy approval, same-connection freshness and observed page effects remain unproven;
-protected-main approval, central workflows and release gates still apply.
-
-##### Receipt descendant adoption: 16:17 UTC
-
-This checkpoint supersedes the earlier descendant-adoption next action for the
-following four owners. They now include the pointer receipt repair by ordinary
-parent merges; no history was rewritten and no protected merge occurred.
-
-| Owner | Published exact head | Parent | Fresh CI |
+| Buyer-visible gap | Evidence at this cut | Canonical owner / next acceptance | State |
 | --- | --- | --- | --- |
-| #259 | `180c168ecbdcd5eb7a4ad14ab4a53e8670646bf7` | #258 `5417ce32` | `34043222194` queued |
-| #260 | `3807aabeb22f9622610c3c8d504d1c686d25d896` | #259 `180c168e` | `34043664230` queued |
-| #261 | `ba100fbc39e1ac4f10ee4faade38418551bb8298` | #260 `3807aabe` | `34044193429` queued |
-| #277 | `121578a43adda12221a9ca9ab8ada0a4fd03efef` | #261 `ba100fbc` | `34044857929` queued |
-
-Canonical regression replays `6999252e`, `8f6b5a0c`, `c03e0dcc` and `3734a874`
-each reproduced a replacement connection consuming an original pending pointer
-command. #259's merge also exposed a raw-response fixture compile failure;
-`d1fd06bf` adopted the existing sealed reader and retained its returned connection
-for the subsequent event. Descendants preserve that repair and the stronger
-foreign-success/error rejection with original-response recovery. Their respective
-focused suites passed 16, 18, 21 and 23 tests; full local Rust 1.97.1 gates passed.
-#259–#261 passed 142 Python contracts each; #277 passed 144 Python contracts.
-
-Exact numerical function/line/region/branch coverage is
-1140/11847/15133/1332, 1154/11975/15318/1334, 1173/12110/15515/1334 and
-1222/12821/16447/1418 respectively, each 100%. The unstable branch-instrumentation
-warning remains visible. #259's Manifest V3 run `34043222184` is also queued.
-Independent read-only reviews found no actionable preservation findings; they are
-not counted GitHub approvals. Actual Edge screenshots verified the published PR
-evidence, not product-browser acceptance. All four source writers are released.
-
-Document-epoch, origin-validation and subscription-specific production/tests remain
-unchanged. Subscription deadlines still fail before registration; proven no-byte
-failures retire only that request, while uncertain writes preserve pending state.
-The next bounded source work is subscription response provenance in #277, followed
-by content-aware descendant integration. Preserve result projection before state
-consumption. Status receipt prerequisites and pointer outbound session authority
-remain separate gaps. The protected-main asset-preservation finding also remains:
-the active source stack lacks this baseline file, so pairwise CI cannot prove safe
-main integration. Browser ownership, event provenance, causality and release gates
-remain unfulfilled. This documentation update requires its own checks and visual
-inspection; source coverage does not validate the documentation revision.
-
-##### Subscription response repair: 16:37 UTC
-
-This checkpoint supersedes the subscription response provenance next action above.
-#277 is published at `46ae62aa31e35c702cd61c16322d05c7a9c35da1`, with unchanged
-#261 parent `ba100fbc39e1ac4f10ee4faade38418551bb8298`. On baseline `122ca139`,
-both success and error replies from a second connection to the same listener/session
-consumed the original subscription (0/2 regressions passed). Repair `8b1508c8`
-reuses private sender registration and the sealed receiving-message consumer;
-replacement replies now leave both requests outstanding and the original reply
-completes only its own request. Required result projection still precedes state
-consumption; deadline and proven no-write safeguards remain unchanged.
-
-All 14 focused subscription tests, full Rust 1.97.1 gates, 144 Python contracts,
-compileall, CodeGraph and diff checks pass locally. Numerical production coverage
-is 1222/12824/16453/1418 functions/lines/regions/branches, each 100%; artifact SHA-256
-is `68183d40d92663f604db48342c0eee0428e18836ed008fee1268e6cf14b5d60a`.
-The unstable branch-measurement warning remains. Exact-head CI `34045953423` is
-queued. Independent source and documentation reviews found no actionable findings;
-they are not counted approval. Source writer `5560556181` is released.
-
-Current-head visual inspection remains pending: the actual browser tool reports
-the Mac is locked and automatic unlock failed. Earlier screenshots do not validate
-this repair or the present baseline update, and are not product-browser acceptance.
-The next safe source queue is canonical outbound session binding, same-connection
-freshness, status receipt prerequisites and protected-main asset preservation,
-with owner-first dependency integration rather than a cyclic descendant merge.
-Navigation-event authenticity and causal page effects remain unproven. No workflow,
-approval, protected merge, tag or release changed. This baseline revision requires
-its own documentation tests, hosted checks and visual inspection.
-
-#### Session repair and child adoption: 13:35 UTC
-
-This update supersedes the 12:28 source/adoption and visual-boundary claims below.
-The complete five-page live inventory read at 13:29 UTC still found **125 open pull requests: 12 Ready/non-draft and 113 Draft; 13 open non-PR issues**.
-All 12 Ready PRs were BLOCKED, with only #147's unresolved thread
-`PRRT_kwDOTulPlM6coZwc`. Subsequent REST reads confirmed 13 non-PR issues, no releases
-or tags, and unchanged protected main `87c4daa1830bac5a5228b6036752ad5633232085`.
-
-Parent #264 is now `6f331a5b220349a1aaa1b1841d5e8ec027b9ad49`, Draft on #263
-`3f22de94b63da83eaa8b5b1270912b21a3ecd006`. Its original actual-socket regression
-was executed and reproduced foreign-session dispatch before the canonical session
-mapping guard repaired it. Unknown, retired and mismatched mappings reject without
-creating state, registering correlation or emitting command bytes. The fixture now
-waits for seed-Pong setup before simulating disconnection; the diagnostic unit-copy
-coverage case is retained. Full Rust 1.97.1 gates, 145 Python contracts and exact
-1282/13455/17136/1440 production functions/lines/regions/branches each pass at 100%.
-Artifact SHA-256 is `b85acd79c3e3a3a49561b9148b6aafadc26cd292a542ec74ef2b523e88c42eaa`.
-Current CI `34035628391` and MV3 `34035628337` remain queued, not hosted GREEN.
-The source writer released in comment `5559331330`; actual Edge visual inspection
-of its published body and Checks page was completed on this head.
-
-Child #265 now publishes `7147893c96ca95c9b5b275d8011c5bfe99aab065` and actually
-adopts that parent. Original parent regression replay `98621adf` first reproduced
-the same socket failure on the child (0 passed, 1 failed). An ordinary merge then
-preserved both histories, admitted-node/current-document authority, typed dispatch,
-deadline and ambiguous-write safeguards, all four previously restored postcondition
-tests, and both Proposed ADR refinements. All 16 focused regressions, full Rust
-1.97.1 gates, 145 Python contracts, compileall, CodeGraph and diff checks pass.
-Exact pinned-nightly coverage is 1296/13601/17303/1442, each 100%; artifact SHA-256
-is `480d6d3ca492565e8a04bfce4b0c135dffbff276532d129c410674838e8fec2f`.
-CI `34036335342` and MV3 `34036335505` remain queued. Fresh actual Edge visual inspection
-verified the published body and Checks page at this head: readable wrapping, no
-observed clipping or overlap, and exact job IDs matching the API. This is GitHub
-presentation evidence, not product-browser acceptance. Writer `5559497861` released
-after normal publication and readback. Independent preservation review found no
-actionable finding; it is not counted approval.
-
-These are local parent/child acceptance results, not protected-main delivery.
-The branch-instrumentation warning, #195/#279 foundation prerequisite, counted-review
-rule, browser authentication, action authorization, pointer-reply connection binding,
-navigation causality and release gates remain separate. Earlier revision measurements
-and screenshots are historical. This new documentation revision still requires its
-own post-publication visual inspection; prior #238 `112ba13a` inspection is not transferred.
-
-#### Text-input adoption and review correction: 13:52 UTC
+| Complete governed presentation identity in real Chromium | #292 remains the browser-runtime acceptance issue. Ready PR #299 is exact `f062cacf1097efd2305e881041a3bf87a660c923`; native CI `34330546178` succeeds, while Manifest V3 Compatibility `34330546195` fails before browser acceptance can pass. | OriginWeave Browser Session/presentation domain consumes #148 diagnostic authority and #212/.github least-privilege Chromium sandbox contracts only after their owner generations are immutable. Re-run three independent pinned-Chromium trials through ambient observation → apply → page-observed target → native interaction/outcome → reset → observed baseline → session/profile cleanup. | **OPEN** |
+| Trusted docs-only CI classifier before workflow suppression | Ready #287 is exact `3975daf48e01a5e9d1cf9fb104a3be1aa03b0402` directly on protected main. Native CI `34329801487`, Security Scan `34329801483`, and Semgrep `34329801598` succeed; CodeQL `34329801477` remains failed at the central verdict path, not at an observed classifier-source finding. | #287 must receive the required central verdict and independent review, reach protected main normally, then #279 may activate workflow consumption from the protected generation. | **OPEN** |
+| One governed Rust stable baseline across product contracts and workflow materialization | Draft #308 is exact `0cf4275d364f529eb3c23dbbfec5ce20113db47e`; it preserves the Rust `1.98.1` successor delta without editing `.github/**`. Native CI is skipped by Draft admission; Security Scan and Semgrep succeed; central CodeQL remains a separate owner defect. | #279 owns OriginWeave workflow materialization. `.github` owns reusable CodeQL publication/wake mechanics. After owner integration, rerun repository contracts, rustfmt, locked workspace tests, strict Clippy, rustdoc, exact production coverage and all central required checks on one unchanged head. | **OPEN** |
+| Buyer decision surface remains usable while evidence history stays immutable | The previous baseline blob is 186,500 bytes and includes repeated dated checkpoints before the commercial gap surface. This successor preserves that exact blob under `docs/evidence/` and keeps this path bounded by regression. | #238 documentation owner: dated observation receipts and the buyer matrix must remain reachable near the top; detailed runbooks belong in immutable evidence/traceability receipts. | **IN REPAIR** |
+| Protected merge/release evidence | Active rules require counted approval and central workflows. Current Release inventory is empty. | Normal protected-main merge only after exact-head checks/reviews. Release-ready head must then re-read tag/package state and produce version/CHANGELOG, signed artifact, SBOM, provenance, reproducibility and rollback evidence before immutable release. | **OPEN** |
 
-Published #266 `eb6c236ff2f4a58b807a2f2c914bd1ddb6079fb3` adopts #265 `7147893c`
-through ordinary merge `5a722867`, preserving original child `cc9980c0` and its unchanged
-text-input source/eight tests. Actual inherited socket RED `18a64573` failed before
-adoption. All 24 focused regressions, complete Rust 1.97.1 gates, 145 Python contracts,
-compileall, CodeGraph and diff checks pass. Exact pinned-nightly production coverage is
-1309/13749/17512/1452, each 100%; SHA-256
-`596097f5545490f6a4bd29a7b718a61aeb43f0ef738e02a61f372232ecd64543`.
-CI `34037184023` and MV3 `34037183844` remain queued. Actual Edge inspection of the
-published body and Checks page confirms the exact head, Draft state and queued jobs,
-with readable wrapping and no observed clipping or overlap. Writer `5559594727` is
-released. This is not product-browser acceptance or protected delivery. #267 already
-owns typed text transport and needs separate current-parent adoption and dispatch review;
-#266 itself still provides construction, not text dispatch or observed action success.
+No row above authorizes a local copy of Wardnet, EgressWeave, Keyverse, contextual-orchestrator, Context Fabric, `.github` workflow source, or another repository's mutable PR head. Browser-domain truth stays in OriginWeave; external capabilities enter through released/versioned contracts and ACLs.
 
-The later complete queue read confirmed the canonical inventory above and all Ready
-roots remained BLOCKED, but found two new #238 review threads in addition to #147: `PRRT_kwDOTulPlM6fsIep` and
-`PRRT_kwDOTulPlM6fsIet`. Both findings were verified against `ca8f8475`: the supplement
-presented predecessor source claims as current, and the inventory assertion selected
-historical counts by formatting. Regression `6fe2af69` observed both failures. The
-repair explicitly labels the supplement historical and bounds current inventory checks
-to this latest cut, paired with the dedicated current CHANGELOG entry. A mutation test
-proves that a changed latest count cannot be hidden by unchanged historical counts.
-Thread resolution requires published repair evidence; this documentation revision still
-needs its own post-publication visual inspection and exact-head hosted checks.
+## Acceptance order
 
-#### Current executable queue and lineage: 14:18 UTC
+1. Repair canonical prerequisites in their owner repositories or bounded contexts; do not patch around them in an OriginWeave leaf.
+2. Adopt owner deltas by ordinary merge/restack without force-push or destructive rebase and re-read the exact resulting head/base.
+3. Run repository contracts and full Rust quality gates on that exact head. Draft/skipped checks are not GREEN.
+4. For browser claims, execute realistic pinned Chromium. Success requires browser-observed post-conditions, navigation/session provenance, crash/cleanup handling and all required independent trials; command ACK is insufficient.
+5. Satisfy counted review, required central workflows and thread resolution before protected-main integration.
+6. Only a protected, release-ready exact head may proceed to signed artifact, SBOM/provenance, reproducibility, rollback, tag/package and immutable release verification.
 
-Historical Ready roots: #37, #50, #166, #219, #220, #229, #238, #240, #272, #274, #285, #287.
+## Evidence and history index
 
-The fresh complete inventory confirms the canonical counts above. These are Ready
-review candidates, not merge-ready approvals: all remain BLOCKED. Re-fetch each
-head, base, checks, findings and counted approval before acting; do not substitute
-the historical root list below. #238's moving head remains live metadata rather than
-a self-referential SHA in this document.
+- [Historical product/technical gap dossier through the 2026-09-09 `16098a0...` generation](evidence/product-technical-gap-baseline-through-2026-09-09.md) — byte-for-byte preservation of the former baseline, including its previous verified cuts, historical PR heads, checks, measurements and runbooks.
+- [Current decision-surface navigation traceability](traceability/product-gap-baseline-navigation.md) — problem, constraints, alternatives, decision, risks and acceptance contract for separating buyer navigation from evidence history.
+- [Product roadmap](product-roadmap.md) — planned delivery sequence; plans do not establish shipped state.
+- [PRD](PRD.md) and [TRD](TRD.md) — product intent and technical design with maturity labels.
+- [Test strategy](TEST_STRATEGY.md), [operability](OPERABILITY.md), [threat model](THREAT_MODEL.md), and [release/rollback contract](RELEASE_AND_ROLLBACK.md) — acceptance and operating controls.
 
-Current foundation #195 `63997bcf555e2c5c8e91ba287734ffba3837a1b7` remains Draft on
-`6922dd98779e8f8aad132a3b1f563d7ba6e6d070`. The earlier `48eb2d` observations are
-historical, not current lineage. #279's workflow-generation prerequisite and #212's
-sandbox-helper boundary still require their own evidence and must not be bypassed.
+### Historical contract anchors
 
-Published transport #267 `4435ce5f561ca069c1844a1a5bd9b603505e25f7` adopts #266
-`eb6c236ff2f4a58b807a2f2c914bd1ddb6079fb3` through ordinary merge `d903cf6b`, preserving
-both histories. Actual REDs `2b960002`, `e2b49e68` and `10131eb7` preceded repairs for
-invalid deadlines, no-write preflight retention and foreign-session dispatch. The
-implementation reuses canonical timeout/session checks and sealed typed dispatch;
-ambiguous writes remain pending. Ten focused socket tests, full Rust 1.97.1 gates,
-145 Python contracts and exact production coverage 1317/13832/17596/1456 each at 100%
-pass. Coverage SHA-256 is
-`065e535fd49c34699b35d53b4cc6b09f22b3dff506f07eb5549c21b73074bdc8`.
-Actual Edge inspection verified its published body and Checks page at that head,
-with readable wrapping and no observed clipping or overlap. CI `34038399969` remains
-queued; this is not hosted acceptance, counted approval or product-browser evidence.
-Source writer `5559716889` is released.
+The archived dossier retains the full **Observed snapshot: 2026-08-29** and all later dated implementation receipts. The short anchors below exist only so current readers can understand long-lived product boundaries without traversing that dossier. They are not a claim that those historical heads are current.
 
-Execute the verified #238 root/lineage review repairs first, then adopt #267 into
-existing response owner #268 `8d4027e40b790d28d866051ba741db12927ec22c`. That consumer
-still uses removed generic correlation; migrate to the parent's distinct text family
-without restoring generic routing, weakening malformed-response retention, or treating
-response admission as observed action success. Received-message provenance remains a
-separate unproven boundary. Keep source and documentation writers separate. This
-documentation revision needs its own publication, visual inspection and hosted checks.
-
-#### Prior observation: 12:28 UTC
-
-All current-state wording in this prior observation is scoped to its recorded time.
-
-Observed through (UTC): `2026-09-06T12:28:16Z`. This cut supersedes volatile claims in the prior observation section below; those earlier exact-head measurements remain historical rather than transferable acceptance evidence. The complete queue read finished at `12:28:16 UTC`, after #265 publication and writer release. The new parent #264 regression is independently owned and is not yet adopted by #265.
-
-The inventory remains **125 open pull requests: 12 Ready/non-draft and 113 Draft; 13 open non-PR issues**. Protected main remains `87c4daa1830bac5a5228b6036752ad5633232085`; release and tag inventories are both empty. Ready-root source heads other than this document's moving branch are unchanged. GraphQL access recovered after the earlier rate-limit failure. GraphQL thread resolution was refreshed for all 125 open PRs alongside exact heads, bases, check rollups and formal reviews in five pages, finishing at `12:28:16 UTC`; all nested histories fit their 100-item pages, with no remaining pagination flags. #147 retains the sole unresolved thread `PRRT_kwDOTulPlM6coZwc`. #166/#220 retain `CHANGES_REQUESTED`, and the other Ready roots retain `REVIEW_REQUIRED`; all twelve are `BLOCKED`. The earlier successful GraphQL cut at `10:00:47 UTC`, subsequent failed inventory/minimal probe and complete 375-read REST fallback at `10:12:37 UTC` remain historical diagnostics, not the current freshness boundary. Active ruleset `18156473` still requires a counted approval and seven central workflows, while the collaborator inventory still contains only the author. #219 has a same-head formal bot approval, but its freshly verified GitHub decision remains `REVIEW_REQUIRED`; an uncounted review does not satisfy the gate. None of this authorizes a bypass or substitutes for an eligible counted approval.
-
-The restored foundation changes the next executable queue. #195 is Draft at `63997bcf555e2c5c8e91ba287734ffba3837a1b7`, on `6922dd98779e8f8aad132a3b1f563d7ba6e6d070`. Its owner restored product/evidence assets lost by historical whole-tree repair `5c111d0db6c363f9d1786c21cc01c5c7398007bd`; this is a content-recovery boundary, not permission to copy an old tree over later work. Fresh hosted CI `34013251657` fails in repository contracts while exact production coverage and MV3 `34013251651` succeed. The exercised failures compare inherited workflow concurrency and `nightly-2026-08-01` against the protected repository-scoped identity and `nightly-2026-08-18`. #279 owns current-workflow reconstruction; weakening the restored tests does not repair the generation mismatch. #242 remains at `2d0e9f69df9ade21d8e8e3d807c3ff644d83b310`, with a stale pre-recovery #195 base `48eb2d23009c1c804520dd5efcd0d4d072aacef1`. Its old CI success is not recovered-foundation acceptance. After the owner prerequisite, adopt the verified foundation content-aware and non-destructively before regenerating descendant checks.
-
-The owner has already moved #255 directly onto #252 so the valid #253/#254 deltas and connection-bound repair travel together. #253/#254 remain open, not silently discarded. Current native Rust and coverage checks succeed on each of these exact heads; earlier local coverage counts do not validate these newer heads:
-
-| PR | Current head | Exact native CI |
-| --- | --- | --- |
-| #255 | `6865faa8185c5fe6a8b6aaba9535d774575b0061` | `34003175242` success |
-| #256 | `881c7f09ee9161ce8664dd75226938ecf60b85e5` | `34003259914` success |
-| #257 | `8f1507346f65798a6bf4eaf370d65a2d406a6f44` | `34003269343` success |
-| #258 | `b8eaa97f7a417c79250b6b760ff214ac03d39b8a` | `34003285350` success |
-| #259 | `91d95423cf31947f691db5ebbd3072c481d86542` | `34003299669` success |
-| #260 | `e5228396be8d9faade44a30aed704cacbeb91b46` | `34003322381` success |
-| #261 | `572fc4224ddc09c010bd9ccf100764076899495a` | `34003334868` success |
-| #277 | `973e34bc24ae9bdd96a50764f2be6c8603eff66c` | `34003345253` success |
-| #263 | `3f22de94b63da83eaa8b5b1270912b21a3ecd006` | `34009256997` success |
-
-#263 now targets actual #277. Its current Rust and exact-coverage jobs `101422055630` / `101422055538` succeed after the owner repaired formatting and exercised the unsubscribe frame-error branch. This is native exact-tree evidence, not protected integration, authenticated browser post-condition, or central review acceptance. The deeper #195 foundation prerequisite still applies to this whole stack.
-
-#288 is now Draft at `fd589cd693946ef1ce2c9270c2dfb6a1087bdfb9`. The independently inspected two-file diff from `0f434bc...` contains documentation-consistency RED `e74acd7...` and the doctoring correction `fd589cd...`, with no production or workflow change. Its sandbox-source and historical mock-fixture boundaries are explicit. Current Strix and Noema succeed, CodeQL fails, and native Rust/coverage/MV3 are skipped; predecessor local 173-test/coverage proof is not current browser execution. The earlier planned documentation repair is done and must not be duplicated.
-
-#148 is now Draft at `0135984f1bc1f68d89d7777f49c4999474105a12`. Current native CI `33990522263` succeeds, but pinned-Chromium MV3 `33990522248` fails. The later bounded diagnostic artifact locates browser-crash failure at session creation; #212 records 0/3 for every real-browser lane after all sandbox-disabling overrides were removed. #43's separate helper-generation MV3 `33866932365` succeeds only on its own exact source. #212 owns sandbox-helper adoption against the current protected workflow, followed by fresh consumer execution; cleanup bookkeeping and leaf success are not transferable runtime evidence.
-
-#150 is Draft at `6ab7fc9166f397f49442243492ae88d6ccad56cf`, ordinarily adopting #148 `0135984f1bc1f68d89d7777f49c4999474105a12` while preserving its combined ordinary-pass teardown waiter. Reused parent regressions first exposed three unsafe browser-launch paths and missing crash-stage diagnostics on the child predecessor. The repair preserves all sandboxed parent launch/cleanup behavior and the child waiter delta. All 253 Python contracts pass on Linux without skips; the corresponding macOS run explicitly skips three Linux pidfd cases. Host/guest tracked manifests match `985411ef6a7d3b812be086b43b6341a96f6b8acace63c0161fa8453aae065bd3`. Complete local Rust gates and 415/3555/4444/476 exact 100% coverage pass. Native CI `34018588292` now succeeds, but MV3 `34018588298` / job `101446865332` fails on that same head. Artifact `9985228944`, digest `sha256:05b6d1cd6092a9e2e18a5003a5a1faeb73b8a8d9b4fd7533f7f24bc938833436`, records all four browser lanes at 0/3, with each browser-crash trial failing at `session_create` and all profiles cleaned. This identifies the failed stage, not the underlying Chrome startup cause. The bounded diagnostic and sandboxed runner/helper follow-up is attached to [#212 comment 5557994915](https://github.com/ContextualWisdomLab/OriginWeave/issues/212#issuecomment-5557994915); no sandbox bypass or unchanged rerun was attempted. #147's ordinary-deadline review thread remains unresolved until this child is actually integrated.
-
-#### Subscription provenance predecessors
-
-The following paragraphs preserve exact predecessor observations. Their remaining-boundary statements and queued checks describe those revisions, not the newer original-registry repair below.
-
-#264 remains Draft on actual #263 `3f22de94b63da83eaa8b5b1270912b21a3ecd006`. The preceding published head `43d3b5a3a2b5ce4f51a93d1152a0ee82620f4f3e` retained private original-command identity after test-first `73f11de2232060ac7680e188db88ef7609123296` and the two-registry collision regression. Its five admission tests, 144 Python contracts and 1273/13294/16963/1430 exact 100% local coverage are predecessor evidence; native CI `34020055803` is now cancelled, not current acceptance.
-
-The inbound-provenance repair was first preserved as unpublished checkpoint `bd29f405a6c927b2f7d1c437dccbfb8bcec424f1`, following real crossed-connection RED `918c4ebe`. It reused the existing connection-owned reader and private connection generation to reject foreign success, protocol-error and navigation-event messages before pending-command or document mutation, then proved original-connection recovery. Seven admission tests, 23 focused tests, all 144 Python contracts, full Rust gates and 1273/13311/16973/1432 exact 100% coverage passed on that private tree. Independent read-only review found no actionable defect; it is not counted approval. Those measurements remain attached to that predecessor, not transferred to the later integration below.
-
-A separately active writer advanced the remote through `bc2e69d5`, `a3dfd190`, `c46642e2`, `ed434fc7`, `d8983d50`, `ced211a7`, `7ae9657c`, `7c20907b` and then-current head `8ebcc6131a5dd6bf0b4720c2f1ff8d40d1cc39f8`. The inspected intervening commits contain valid regression and connection-provenance work. [Writer coordination comment 5557964460](https://github.com/ContextualWisdomLab/OriginWeave/pull/264#issuecomment-5557964460) requested task identification, writer acknowledgement and a final exact-head handoff. The private writer claim was released without a competing push; the subsequent integration and release are recorded below. The current caller-supplied registry binding, actual-resend freshness on the same connection, unsubscribe transport/lifetime provenance and real-browser causality remain distinct unproven boundaries. The #195/#279 prerequisite, branch-instrumentation warning and Draft status remain; no workflow, dependency, deadline or gate changed.
-
-A fresh unmodified detached checkout at the earlier remote `8ebcc613` diagnostic cut passed Rust 1.97.1 network/all-targets checking, strict workspace/all-targets/all-features Clippy and all 145 Python contracts. At that cut, the Rust 1.97.1 formatting check fails in five test/support files. The pinned nightly coverage test run completed, but the unchanged numerical verifier failed at lines 13327/13330 and regions 17004/17008; functions 1274/1274 and branches 1434/1434 are exactly covered. Coverage artifact SHA-256 is `f352e3f04bdc3ed1912be7c452c5c426d9caaec7dd1a613b7d3ab9c6486ab69e`. The uncovered regions were the duplicated missing-generation propagation in correlation completion at line 240 and the admission diagnostic at lines 396–397. Independent source review confirmed that the former follows an intent check whose retained constructor always sets the generation; shared validation and a real rejected-event diagnostic/recovery test addressed these causes without exclusions or fabricated production state. The comparison found no reachable authority bypass and identified remote typed-send malformed/unknown/error fixtures, stronger rustdoc and the source contract as deltas to preserve together with private original-connection recovery, unchanged document state, authentic receipt redaction assertions and bounded evidence. Neither private nor remote results establish combined-source acceptance. Earlier native CI `34021674375` was queued at that diagnostic cut; formatting, numerical coverage and hosted execution remain separate gates. The coordination comment recorded the diagnostic and preservation map before shared-source integration or writer acknowledgement occurred.
-
-After the other writer's explicit [release at 52cff969, comment 5558261278](https://github.com/ContextualWisdomLab/OriginWeave/pull/264#issuecomment-5558261278), ordinary content-aware merges preserved both complete contributor histories. Published #264 head `2a9fdc5418b9af10353cdad0f6f6470655bf457d` contains private `bd29f405a6c927b2f7d1c437dccbfb8bcec424f1` and remote `52cff96954c3fec7b8cda5409e4560e970bfcbdf`: both complete lineages are now ancestors. Independent read-only review found no actionable preservation defect. Final-tree Rust 1.97.1 formatting, locked workspace/all-targets/all-features check/test/strict Clippy, strict rustdoc and all 145 Python contracts pass; the workspace run retains all 26 focused socket tests. The unchanged pinned-nightly verifier confirms 1273/13317/16984/1432 functions/lines/regions/branches at exactly 100%, with coverage artifact SHA-256 `f7e71e6fd67723688535053c389b4ff718c4563b403924af462848d3d51b541f`. No test exclusion or quality-gate change was introduced. The normal push and exact remote head were verified, and the integration writer released. Fresh exact-head native CI `34024499232` remains queued; this is local combined-source acceptance, not hosted acceptance, counted approval, protected-main delivery, browser causality or release evidence. The unstable branch-instrumentation warning and all unproven boundaries above remain.
-
-#### Original-registry ownership predecessor
-
-The preceding published #264 head is `10f138f8787d596e8b556fe50c9e4e52bc1295b7`, still Draft on the same #263 base at that cut. Four real-socket REDs at `b3ffeac9` proved original-command send, original-receipt admission, original-event admission and document mutation could use a replacement registry with colliding local identifiers. The document-mutation case did not require matching external context text. The repair retains one opaque core-owned allocation witness through the existing command, binding and subscribed observation, rejecting foreign ownership before correlation/I/O, replay insertion or the common document-mutation sink. Origin binding uses that same sink. Original registry moves and cloned witnesses remain valid; owner destruction does not let a replacement inherit the old identity. The socket receipt regression drops the original registry before creating its replacement, and the existing core diagnostic contract exercises the real dropped-owner mismatch.
-
-The final exact head passes all four new socket regressions, all 145 Python contracts without skips, complete Rust 1.97.1 formatting/check/tests/strict Clippy/rustdoc, compileall, graph and diff checks. Unchanged pinned-nightly coverage is exactly 100% at **1278/13371/17056/1434** functions/lines/regions/branches, artifact SHA-256 `9f5b4942d6a040a83f09593a2b81406c44bc9d2b402fa995a95294ce8e9cbb7e`. Initial `e686b3a0` coverage missed one diagnostic line and three regions in the core unit copy; the genuine error path now exercises that copy. Intermediate `33787617` reached 100% but failed the strict test-lint rule; the final test uses the established error-collection/cardinality assertion style. No exclusion, lint allowance, dependency or production authority relaxation was introduced. Independent read-only review found no actionable authority, lifetime, construction, preservation or documentation finding. Both complete contributor lineages remain ancestors; normal publication, the current PR body and explicit writer release were verified.
-
-Native CI `34026519860`, Rust `101468348712` and coverage `101468348775`, and real MV3 `34026519878` / job `101468347302` now all completed successfully on `10f138f8`. These are terminal-success predecessor evidence, correcting the earlier queued observation without validating a later head. The repair pins the original registry but does not authenticate the registry-to-transport association. Actual-resend freshness was still unfinished on this predecessor; the next section records its separate repair. Unsubscribe lifetime/transport provenance, action causality and end-to-end Chromium product acceptance remain unfinished. ADR 0107 stays Proposed, #195/#279 remain prerequisites, and the unstable branch-instrumentation warning remains. No workflow, gate, merge, tag or release mutation occurred.
-
-#### Command-response freshness predecessor
-
-The preceding published #264 head `805051527cf95e14ba126c9dd3159db86d190224` was Draft on unchanged #263 `3f22de94b63da83eaa8b5b1270912b21a3ecd006`. Actual success-first RED `92fd0b07` and error-first RED `15aea15e` each exposed four repeated-send lifecycles: completed, retired, replacement correlation and unread buffered response. Three additional actual-wire REDs exposed raw/typed mixing and repeated identifiers; a compile-fail contract exposed the pre-upgrade socket alias. The shared established owner enforces strictly increasing typed identifiers and exclusive raw-text/typed mode across all five typed senders. Zero is valid first, out-of-order replies for distinct outstanding commands remain valid, and reader reconstruction/Pong cannot reset history. The consuming socket handoff remains; its nonconsuming alias is removed. This is a stricter local dispatch policy, not a W3C requirement or authenticated browser-session association. The next section records its later teardown refinement; this predecessor's unfinished boundaries and screenshots are not current-head acceptance.
-
-All 22 focused admission/frame tests and the added real public opening-deadline test pass within the complete Rust 1.97.1 workspace verification. Formatting, locked all-target/all-feature check, strict Clippy, warning-denying rustdoc, all workspace tests, all 145 Python contracts, compileall, CodeGraph and diff checks pass. Unchanged pinned-nightly coverage is exactly **1279/13409/17087/1436** functions/lines/regions/branches, artifact SHA-256 `64a62b1e03ee3ed3d62654a3227495e82b7ff93357038ae146d8584c581ac060`. Earlier buffered-fixture BrokenPipe/reset failures and `e6c02cf`'s 17086/17087 region result remain recorded. The latter combined complementary ordinary-library/unit-test gaps rather than unioning source regions; the real public deadline case covers the ordinary copy, retaining the private real revoked-socket test without production or verifier changes. That one-nanosecond case has local passing evidence; portability beyond executed targets remains unproven. Independent read-only review found no actionable finding and does not count as GitHub approval.
-
-Fresh exact-head CI `34029687813`, Rust `101476824185` and coverage `101476824305`, and MV3 `34029687816` / `101476824298` are queued. Post-publication actual Edge visual inspection covered the new PR body and Checks page: exact head/base, Draft state and three queued checks matched the API; desktop screenshots showed readable wrapping with no observed clipping or overlap. GitHub presentation evidence is not OriginWeave product-browser acceptance. Both contributor histories remain ancestors, ordinary publication was verified, and the #264 writer was explicitly released before this documentation-only slice. Registry-to-transport authentication, unsubscribe lifetime/transport provenance, action causality, real browser acceptance, #195/#279 integration, protected delivery and release gates remain open.
-
-#### Subscription teardown ownership and provenance predecessor
-
-At the preceding cut, published #264 was `7fb93e77b800f27187a5c02333298cc31a025bd6`, Draft on unchanged #263
-`3f22de94b63da83eaa8b5b1270912b21a3ecd006`. Lifetime RED `2c45cea8` reproduced event admission
-after borrowed teardown construction. Transport RED `ce6f6fd4` reproduced 91 actual masked bytes
-on a foreign connection and foreign success/error consuming the original pending command. Genuine
-original replies then failed as no longer outstanding; real servers were joined before assertions.
-
-The repair consumes the existing non-cloneable subscription receipt and retains its original
-connection identity through teardown dispatch and acknowledgment. It reuses connection-aware
-correlation and the existing connection-bound reader; no revocation registry, dependency, workflow
-or quality-gate change is added. Receipt reuse is now rejected by the compiler with `E0382`, not
-merely by a fixture setup error. Construction failure also ends local admission availability.
-Previously admitted observations are not retroactively revoked, and acknowledgment does not prove
-event drainage or browser cleanup. Proposed ADR 0107 records these costs and rejected alternatives.
-
-All three transport regressions, eight preserved failure cases, fourteen admission tests and the
-escaped-identifier round trip pass. Full Rust 1.97.1 workspace/all-feature formatting, locked check,
-strict Clippy, warning-denying rustdoc and tests pass, including three network doctests. All 145
-Python contracts, compileall, CodeGraph and diff checks pass. Unchanged pinned coverage is exactly
-**1279/13421/17106/1438** functions/lines/regions/branches; artifact SHA-256
-`277c5bc0a17f16966a8eb388d54e4e172be36356d7dead5bbead9ed53ad8cff6`. The unstable branch warning
-remains. Independent read-only source review found no actionable issue, not counted GitHub approval.
-
-New exact-head CI `34031281825`, Rust `101481131938` and coverage `101481131907`, and MV3
-`34031281812` / `101481131930` are queued, not GREEN. Both contributor histories are preserved,
-normal publication was verified, and source writer `5557964460` was explicitly released before
-this documentation writer began. The required new-head visual inspection remains incomplete:
-actual browser control reports that the Mac is locked and automatic unlock is unavailable. Manual
-unlock was requested; neither the previous #264 screenshots nor the previous #238 Preview validates
-these new heads. Registry-to-browser authentication, action causality, real Chromium acceptance,
-#195/#279 integration, protected delivery and release gates remain open.
-
-At this read, #238 predecessor `81e334aff063290bde3298671cfd20e77c7c1fff` still has queued native
-CI `34030165220` plus central review/security work. This documentation commit needs its own checks;
-neither its earlier 172 local Python passes nor #264's source coverage transfers to the new head.
-
-#### Current pointer integration and independently owned parent repair
-
-Published #265 is `35555d0f8491d4ee95c2e61d1a2aaa2c02a0635c`, still Draft. Ordinary two-parent
-integration `7535d8af` preserves original child `ffa70ee0f499b86ff51837fb95733fd5cf57ff89` and
-parent `7fb93e77b800f27187a5c02333298cc31a025bd6`. Real-socket RED `3f9cdc1a` observed no command
-bytes but one outstanding command after a zero deadline. The repair retains the child's immediate
-admitted-node revalidation and the parent's typed dispatch, local no-write retirement and ambiguous
-write retention. Both use the existing registry identity allocation. Independent review identified
-four silently deleted parent postcondition tests and two obsolete sender calls; all were restored
-or migrated without dropping assertions. The final review found no remaining actionable finding,
-which is not a counted GitHub approval.
-
-All 19 focused tests, complete Rust 1.97.1 formatting/check/tests/strict Clippy/rustdoc, all 145
-Python contracts, compileall, CodeGraph and diff checks pass. The exact-head pinned-nightly rerun
-passes unchanged production coverage at **1293/13567/17273/1440** functions/lines/regions/branches,
-each 100%; artifact SHA-256 is
-`b1374f65538fb6557b4e5d877aa067e183202ed1f56c75be047b3ff2dfdc92a0`. This is local source evidence.
-CI `34032661775`, Rust `101484986934` and coverage `101484987071`, and MV3 `34032661781` /
-`101484986948` remain queued. Normal publication was verified; writer `5559058158` released.
-
-The live parent #264 subsequently advanced to test-only `b4702cd503fa3f721e0d1f44b355563753dac0a2`.
-Its 129-line socket regression targets registry-session A dispatch over transport-session B;
-writer [5559076006](https://github.com/ContextualWisdomLab/OriginWeave/pull/264#issuecomment-5559076006)
-released in [5559140277](https://github.com/ContextualWisdomLab/OriginWeave/pull/264#issuecomment-5559140277)
-without executing the regression or implementing a production repair. Parent CI `34031977586`
-and MV3 `34031977597` remain queued. The available local backend can now execute that regression
-under a new source lease. Adoption waits for verified repair: #265's local GREEN is **not latest-parent acceptance**.
-The PR base snapshot still names `7fb93e77`, so it must not replace the fetched live parent as
-dependency evidence. No parent source was changed during this documentation slice.
-
-Browser access recovered after the earlier locked-screen failures. Fresh actual Edge visual
-inspection of #265's published body and Checks page verified head `35555d0f`, Draft state and the
-three queued checks. The inspected desktop screenshots show readable wrapping with no observed
-clipping or overlap. This replaces the earlier lock blocker only for that inspected presentation;
-the new documentation head still requires its own visual inspection. GitHub presentation is not
-OriginWeave product-browser acceptance, browser-process authentication or navigation causality.
-
-The #238 predecessor `88447b7d43e4a87b48fb8fde33f93b2e3ce8cbc5` still has pending native CI
-`34031539064` and central review/security work. Its 172 local Python passes and #265's source
-coverage do not transfer to this documentation commit. The next executable items are this dated
-evidence update, then an executed parent regression and verified repair under a new writer lease; #195/#279,
-#212 and protected-main/release gates remain prerequisite work.
-
-#### Scheduling and central-owner follow-up
-
-This section retains owner-reported scheduling and central-repository evidence. Those reports are
-not refreshed owner-source or released-consumer acceptance in this bounded documentation slice.
-
-The existing hourly-task owner attempted a prompt-only coordination/evidence improvement, but the scheduler rejected the update with `too_many_active_automations` at the ten-active-task limit. The owner reports that a subsequent read confirmed the old prompt and update timestamp, cadence, enabled state and notification preferences were unchanged. This is a failed update, not an applied scheduling improvement; no duplicate schedule or unrelated-task pause was created.
-
-On #238's earlier `a2600ee8...` head, native CI `34018002926` succeeded. The later `38e5eb44...` cut recorded queued CI `34021771606`; subsequent published `7ca86a2f546adf3a0af3fec41d97675115db384d` still had queued native CI `34022635038` before this refresh, together with central review/security work. At the now-predecessor `8e6284f5f89965983d025ce302d7c99e2f5c38d5`, native CI `34027019757`, Rust `101469694749` and coverage `101469694911` have completed successfully; nine central security/review jobs remain queued. This document's next commit requires new exact-head checks; it cannot inherit those results. Existing central CodeQL dispatch/verdict work remains with `.github#712`. Central Strix #1563 remains unmerged at `eaf9594f7fe8d8e1994349289183d6cbad056579`, reported behind its base: the earlier `13fbb48e...` source repair and owner-reported harness results do not prove consumer recovery. The unrelated #37 local 900-second deadline remains distinct from the #166 completed-report classifier defect and #219/#240 actual gateway/hosted-limit incidents. The active central Noema owner additionally reports source REDs under `.github#1641`; owner reports and comments are not released transport or SDK-compatibility evidence, and no consumer workaround is introduced here.
-
-### Prior observation cut: 2026-09-05
-
-The following cut preserves prior source measurements and diagnostics. Present-tense wording within this dated cut describes its observation time only; use the latest verified cut above and live APIs for the current queue.
-
-Observed at (UTC): `2026-09-05T13:50:14Z` (full REST inventory/review/check refresh and targeted source verification; subsequent bounded diagnostics are recorded below).
-
-GraphQL rejected both the inventory query and a minimal probe with its rate-limit error while the REST endpoints remained available. REST reviews, check runs, and combined statuses were read for all 125 open PRs; truncated output was recovered by querying the exact missing records. New source heads were refreshed separately after pushes. Review-thread resolution cannot be inferred from these REST endpoints: earlier thread-resolution statements below retain their prior observation date and are not refreshed merge evidence. No query failure, queued check, or metadata update relaxes a quality or review gate.
-
-- #37 Strix run `33946242342` / job `101285198427` failed at 13:26 UTC. Artifact `9970303662` (SHA256 `93ce81a9c3060f52a23d0910b5a995a2b491aeb3889a54432d954747a97010ca`) records `Strix run timed out after 900s`, 903 seconds elapsed and exit 124; its current-attempt record is interrupted, not completed. The provider-unavailable wrapper label does not establish a provider outage. Canonical Strix deadline/classification ownership received the exact artifact; no consumer timeout change or blind rerun was made.
-- #148 hosted compatibility run `33962062608` / job `101295542349` failed at 13:33 UTC on `bded4fc9d32e5e047cea99d182aa05ae2cd03bf6`: browser-crash recovery fails 0/3 trials, while ordinary Agent Task and forced-close each pass 3/3 and profiles are cleaned. Artifact `9970408851` (SHA256 `daefc93ff2792cbc82a8937107e7992e91447f56692c4ea7071bc2b39c0faa7b`) initially exposed only `RuntimeError`; the stage was not reconstructible from it. Subsequent owner diagnostics and newer all-sandboxed results are recorded in the latest cut, not backdated into this earlier artifact.
-
-- Protected `main` is `87c4daa1830bac5a5228b6036752ad5633232085` through #286. PR #286 skips repository-native CI jobs for draft pull requests; a skipped job is not passing evidence.
-- The September 5 search also returned 125 PRs (12 Ready, 113 Draft) and 13 non-PR issues. PR #290 was converted back to Draft after current-protected workflow review found that its #245-relative patch would drop protected #286 lifecycle controls; queue movement is not protected-main delivery. PR #248 also returned to Draft for current-parent adoption and complete revalidation. Temporary synthesis #291 has since been externally merged and closed as recorded below; its closure does not establish product acceptance.
-- Ready roots are #37 `1e2f41072854edcdbaf0f9ecf14697a3bfd62195`, #50 `ad87cfea59db711cb29ef90559790ba77e22029f`, #166 `e84a1a2cc82b1c666218efd441da97849f47b8c2`, #219 `65e4315d80137badc0b55e1b9617015beb1db568`, #220 `e545b94e1de499b96b867694f80ac04ad247becd`, #229 `024f63690cf05cfe6f0d4a430f0e18ea8fd2c4d6`, #240 `24930a3a9ee79c0b712ee3df6589b0592eb6e18f`, #272 `b1cae8ad1cbd8eb6992037c830aea30b9aa436b3`, #274 `802d0bdff7536d9ac253305d3e0237b4e4a1789e`, #285 `f455c2cd64b3dd3f027c91d396103792a205ddd0`, and #287 `af83c40dd2990a03064a92ca75430a9cc400f098`. PR #290 is Draft at exact head `ebeefcd534db4324498fdb18046ebc6255ddcdf2`, remains a #245-dependent workflow-owner candidate, and is not a root. PR #238's moving exact head is intentionally omitted from its self-referential document; live PR metadata is authoritative. Their hosted exact-head checks remain non-terminal and none has an eligible exact-head approval, so none is merge-ready.
-- PR #166 exact head `e84a1a2cc82b1c666218efd441da97849f47b8c2` and PR #220 exact head `e545b94e1de499b96b867694f80ac04ad247becd` retain formal `CHANGES_REQUESTED` decisions. REST review records were refreshed; thread-resolution state could not be refreshed because GraphQL rejected the current read. The last successful thread-resolution snapshot was the prior 12:36 UTC sweep, not this REST refresh. Thread resolution alone is not approval or grounds to dismiss a review, and the active counted-approval gate remains unmet.
-- Targeted CI RCA at `2026-09-05T06:45Z` confirmed that #219 Noema run `33925442322` / job `101226941175` and #240 run `33925596923` / job `101227537529` had terminated on gateway HTTP 502 without review verdicts. Neither unchanged exact head had a successor Noema run. One supported failed-job rerun per head created attempt 2, with queued jobs `101264704581` and `101264705575`; Noema attempt 2 is queue-admission evidence, not provider recovery or a review verdict. Successful sampled Noema wrapper runs skipped their review jobs and cannot establish recovery. The CodeQL failures independently record dispatch handoff: #219 central runs `33947047322`, `33947036734`, `33947037072` and #240 runs `33947189162`, `33947189634`, `33947187643` match those exact heads; these central CodeQL dispatches remain queued and were not duplicated. Active Strix runs were left untouched. Revisit these exact attempts after a terminal result; do not create repeated retries while they remain queued.
-- A fresh repository Actions query at `2026-09-05T06:12:27Z` returned **115 queued workflow runs**; the newest sampled run was #238 CI `33949183271` on predecessor head `bb2f18570f319a763acc24d7c4206ec9d0a11a29`. This is runner-admission/backlog evidence, not a code failure or passing check; rerunning unchanged heads would only add duplicate queue load. The moving PR's exact current head remains live-metadata-only to avoid a non-convergent self-reference.
-- Active organization ruleset `18156473` (`CWL Central required workflows`) applies to the default branch. It requires one approving review, dismissal of stale approvals after pushes, resolved review threads, extra approval for unattributed changes, and **7 central required workflows**: `opencode-review`, `pr-review-merge-scheduler`, `security-scan`, `strix`, `sast-semgrep`, `noema-review`, and `codeql-pr`. Current `ContextualWisdomLab/.github` consolidates OSV and Scorecard PR scanning into the required `security-scan.yml`; standalone `osv-scanner-pr.yml` and `scorecard-pr.yml` are no longer ruleset entries. Administrative bypass capability is not authorization to use it.
-- PR #284 head `61bcf88c960c6c437ccd29b3fbb73cd4325f9e5a` reached `main` through rule-suite `3948421709`, whose live result is **`result: bypass`** with actor `seonghobae`. The required non-author approval and workflows were not proven before integration. The earlier pending post-merge state is historical: fresh native Rust and production-coverage checks now succeed on `87c4daa1830bac5a5228b6036752ad5633232085`. This is a governance incident tracked by #215, not a policy-compliant merge or evidence that later checks can retroactively authorize it.
-- PR #285 is Ready at exact head `f455c2cd64b3dd3f027c91d396103792a205ddd0` on current protected main. It addresses #284's three post-merge review findings without changing `.github/**` and adds a dedicated regression contract. All 154 repository contracts, the full Rust gates, and exact 100% local coverage pass. Its Ready transition materialized repository-native CI `33930234387`: Rust contracts `101207153048` and Production coverage `101207153244` succeeded by `2026-09-05T03:16:29Z`; native CI success does not replace the seven required central workflows. The Ready event did not materialize fresh central required-workflow runs. Targeted RCA at `2026-09-05T06:53Z` verified that the previously cancelled Security Scan `33924016851`, SAST Semgrep `33924016903`, and CodeQL PR `33924016883` still bind the current head and protected base, with no duplicate central CodeQL dispatch. The convenience CLI failed on a cross-repository workflow lookup before rerunning anything; the supported run-ID REST route then created attempt 2 for all three, with queued scope/language jobs `101265706025`, `101265708882`, and `101265710202`. These are verified replay admissions, not terminal scan results or proof that Ready-event materialization is repaired. Remaining required review evidence and the absent eligible approval still block merge; toggling Draft or creating a no-op commit is not an acceptable substitute.
-- Issue #279 owns the remaining documentation-CI partitioning gap. PR #287 is Ready at exact head `af83c40dd2990a03064a92ca75430a9cc400f098` on current protected main; it is the workflow-independent classifier foundation, changes no `.github/**` path, and is the prerequisite for an authorized workflow owner. Its predecessor failed its own release-record contract; the current head binds Git rename/copy similarity to blob identity, and its focused and full local suites plus exact 100% Rust coverage pass. Exact native CI, Semgrep and scoped Security Scan have now completed as described next; remaining required review/CodeQL evidence and eligible approval are not complete. PR #282 is Draft at exact head `b64e0708584beff3fb54acf226cb3e667773e473` on current protected main. Its latest workflow/test delta removes `converted_to_draft` and `closed` from the CI event list, removes #286's closed-event guard, and makes the repository contract assert those protections are absent. Exact review `5120043505` therefore requires the authorized #279 owner to reconstruct the valid trusted-base classifier/contract partition while preserving #286's lifecycle and closed+Draft fail-closed controls; Draft checks do not transfer as GREEN.
-- #287 CodeQL handoff RCA at `2026-09-05T08:34Z` confirms that exact native CI `33931806137` passed, Semgrep `33931806165` executed its scan successfully, and Security Scan `33931806226` executed Scorecard/Trivy while scope-skipping gitleaks/OSV/dependency-review. CodeQL wrapper `33931806139` failed intentionally after its three language jobs recorded successful dispatch and a pending verdict, not a failed terminal scan. The cross-repository convenience CLI returned 404; direct REST job logs identify the handoff. Central `.github` dispatches `33954721186`, `33955024697` and `33955164029` have display titles identifying the exact OriginWeave target head, but their immutable run heads are central-owner revisions (`71dd84d...` / `27d7331...`), not the product source head. All three remain queued and were not duplicated; acceptance still requires validated target checkout, an authenticated terminal verdict and the original-job replay. OpenCode is queued, Noema/Strix remain in progress, and no eligible formal approval exists.
-- PR #290 is Draft at exact head `ebeefcd534db4324498fdb18046ebc6255ddcdf2` on #245 exact `a769f484e2c110e0523b3b28cd21573f43867562`. Its workflow/repository/PR-number concurrency isolation and manual-dispatch non-cancellation intent are valid, but its predecessor-relative MV3 patch would remove protected #286 `converted_to_draft`/`closed` lifecycle events and the closed-event job guard. Exact review `5120039692` and canonical-owner issue #212 comment `5549868655` require reconstruction on the current protected MV3 workflow; the scheduled product writer did not mutate `.github/**`.
-- PR #283 is Draft at exact head `c904300a6a1bda83af24f84d586f1c5f6a6491aa`, stacked on predecessor #282 head `b54a5856d8201911f05d69622f0d5594a371adf0` rather than current #282 exact `b64e0708584beff3fb54acf226cb3e667773e473`. It must adopt the corrected current parent before its compare can become current evidence. The child remains one prose-only doctoring canary; Draft-hosted jobs are skipped, so no trigger-shape GREEN is claimed for #279. Runner admission is a separate condition.
-- PR #288 is Draft at exact head `0f434bc29d468127412366b6864b733b40b83c4d` on current protected main. Predecessor `39e36256651f62940ec3ca6149067f0cfcb2285a` had 172 passing Python contracts; that evidence is historical. The subsequent test-first sandbox contract and doctoring change led to the ordinary launch override removal, followed by restoration of the original bounded ChromeDriver diagnostic. Direct cumulative comparison from `99fea8985b4649c444ecc72c11e72c0693b17f5b` to current `0f434bc...` confirms one net production line removed: `--no-sandbox`. Fresh isolated verification confirms all 173 Python contracts pass on that unchanged head, including the source contract for both ordinary and Agent Task launch paths. Complete Rust gates and numeric 100% coverage also pass: 521 functions, 4420 lines, 5349 regions, 654 branches; the unstable branch-option warning remains. The doctoring problem paragraph still describes the historical override as current, so its existing author owns a bounded documentation/test correction after this writer releases. Unit-fixture output is not executed browser evidence. #212 still owns authorized workflow activation and fresh sandboxed pinned-Chromium compatibility/Agent Task trials; no workflow, sandbox-helper configuration, or hosted replay was changed here.
-- Issue #28 remains the P0 governed-browser integration target. PR #261 is Draft at exact head `934eb7d37568b439c442ffe1d1f6a9c8f8ed58a0`; #261 now includes actual #260 `2c5049aff97a90958e8262b1d403bdcbd64a1e8b` by normal merge. Its origin-binding implementation and tests remain byte-identical to predecessor `323ac9e147691e9f6572711f5a748e13f1036624`. Native release-contract discovery reproduced zero tests before parent adoption and one passing test after it. Fresh verification passed 18 focused loopback tests, 142 Python contracts, full Rust gates, and numeric 100% coverage: 1173 functions, 12104 lines, 15511 regions, 1334 branches. Exact CI `33967462582` is queued; no previous check is transferred.
-- Repair PR #277 is Draft at exact head `117f6414e8a6db46eb2b32f4ebae85cf2a208371`, based on actual #261 `934eb7d37568b439c442ffe1d1f6a9c8f8ed58a0`. Parent adoption `3c0484174eeda0703492ba76b530be125e3e99dd` preserved its two subscription production files, three test files, and Proposed ADR, with 16 focused tests, 142 Python contracts, full Rust gates, and numeric 100% coverage at 1220 functions, 12774 lines, 16397 regions, 1418 branches. The subsequent deadline repair reused the existing frame validator before registration: its RED observed no subscription bytes but two outstanding commands where only one pre-existing command should remain. Zero and over-limit deadlines now preserve that command and leave the rejected identifier reusable. Post-registration frame-failure retention remains conservative, including the separately tested repeated-mask rejection. Fresh current-tree verification passed 11 focused tests, 142 Python contracts, complete Rust gates, and numeric 100% coverage: 1221 functions, 12781 lines, 16404 regions, 1418 branches. The actual typed family is `NavigationCommittedSubscription`, not the historical proposed `SessionSubscribe` name. The response parser and authority boundaries are unchanged; exact CI `33968278171` is queued, and `warning: --branch option is unstable` remains. Stale conflicting #262 is not superseded, and #263+ are not retargeted or promoted until current-parent hosted acceptance and complete successor evidence exist.
-- Stacked PRs #178 and #85 were repaired without force pushes at exact heads `640e594dbc0a64f251a0f28b8e80943f53337e40` and `c9adea680d6186f1842a280e248ce55da4f1305b`. Parent-relative deltas now preserve the current MV3 diagnostic and extension-authority contracts; local full suites and exact 100% coverage passed, while hosted exact-head acceptance remains independently required.
-- PR #70 is Draft at exact head `77eb0f2ee71783e06171784b7173c0b4cd530e61`; it targets protected main but has not adopted current main `87c4daa1830bac5a5228b6036752ad5633232085`. Its focused Agent Task contract proves the pre-repair `--no-sandbox` launch as source RED and the branch removes that argument only from the Agent Task lane. Sandbox-enabled pinned-Chromium E2E and repository-wide GREEN remain unproven while its exact-head browser/CI/security runs are non-passing.
-- DDD/MCP repair #272 is Ready at exact head `b1cae8ad1cbd8eb6992037c830aea30b9aa436b3` on current protected main. Its exact-head hosted acceptance remains incomplete and no eligible approval exists. Documentation child PR #273 is Draft at exact head `e5c8fcb66bf644dfa750bb1b40ba3d600cb7805a` on predecessor #272 head `fe124e447cad3f679e22337fb6fbdfd135ab3652`, so it must adopt the parent only after #272's current head completes its gates. Both remain active-PR evidence.
-- PR #229 is Ready at exact head `024f63690cf05cfe6f0d4a430f0e18ea8fd2c4d6` on current protected main. Its 157 Python contracts, full Rust gates, and exact 100% local coverage pass, while hosted exact-head acceptance remains incomplete and no eligible approval exists. The presentation-identity work retains the boundary that Chromium application and page-observed effectiveness remain separate adapter/browser-E2E work.
-- PR #281 remains Draft at exact head `adaca6427d68f550b39293a69b7c733430d1c385` on canonical MV3 parent #43 `6f3134d18d3118aab33d28048671dc71a5f47b77`. Its child delta is diagnostic-evidence contract/doctoring only; inherited #43 implementation redacts browser/page-derived HTTP and WebDriver protocol data, startup exception details, returned capability values, DOM-dataset values, and click post-condition text before CI/audit evidence. Exact-head CI and Manifest V3 Compatibility succeeded and review threads are resolved, but parent-first workflow-ownership verification remains open, so the child stays Draft.
-- PR #43 is Draft at exact head `6f3134d18d3118aab33d28048671dc71a5f47b77`; it targets protected main but has not adopted current main `87c4daa1830bac5a5228b6036752ad5633232085`. The branch now installs and configures the pinned `chrome_sandbox`; exact-head hosted verification remains required, and restoring `--no-sandbox` is not acceptable.
-- PR #274 remains the bounded GitHub Pages source/README/CHANGELOG public-surface delta at exact head `802d0bdff7536d9ac253305d3e0237b4e4a1789e`, with a repository regression contract for the exact badge target, pre-alpha status, active-PR non-promotion statement, and publication boundary. Source presence is not publication evidence; live HTTPS publication and navigation remain required after the authorized Pages configuration/deployment path.
-- PR #37 is Ready at exact head `1e2f41072854edcdbaf0f9ecf14697a3bfd62195`. The head repairs the Rust 1.97.1 formatting failure in its segmented Content-Length regression; formatting, locked workspace check/test, strict Clippy, rustdoc, all 167 Python contracts, and pinned-nightly production function/line/region/branch coverage pass locally at 100%. The reader returns after the exact declared bytes, already-buffered surplus remains fail-closed, and valid self-delimited content does not require TLS EOF. Native Rust contracts and coverage have succeeded on this head, but central workflow evidence remains incomplete and no eligible approval exists.
-- PR #50 is Ready at exact head `ad87cfea59db711cb29ef90559790ba77e22029f` on protected main. At predecessor `e981ac45d0bfcd3906fc64dae5f4490edf39f9e5`, seven focused tests passed but pinned Rust formatting failed on the default-port regression. The formatter-only repair preserves all default HTTP/HTTPS and explicit-port assertions and every production source. Fresh independent verification passed all seven fresh-resolution tests and 152 Python contracts, compileall and complete Rust 1.97.1 format/check/workspace-test/strict-Clippy/rustdoc gates. Numerical coverage is 530 functions, 4509 lines, 5441 regions and 660 branches at 100%, with the unstable branch-option warning retained. Current CI `33964793228` and compatibility `33964793282` remain non-passing queue evidence. Earlier `2bd85188...` documentation and `ddbefc91...` origin-port repairs remain historical: an origin-approved IP could be paired with a different service port; the planner now binds the socket port to the effective scheme-host-port origin before I/O. ADR 0005 and doctoring retain trusted monotonic time and revalidation before socket I/O. Local admission tests do not prove an HTTP/TLS exchange, browser navigation, hosted acceptance or required approval.
-- PR #269 is Draft at exact head `7854394266d3f292e779193c01413a34f6798d7c`, stacked on #268 exact `8d4027e40b790d28d866051ba741db12927ec22c`. It adds only a fixed, product-owned `script.callFunction` text-value observation for the exact admitted current node; it performs no browser I/O and proves no post-condition. Its new documentation boundary contract and all 140 Python contracts pass locally, while hosted exact-head CI is queued and the parent-first dependency keeps it Draft.
-- PR #270 is Draft at exact head `191a14535219ea8033777fa4c970efb281b62418`, now non-force synchronized onto #269 exact `7854394266d3f292e779193c01413a34f6798d7c`. Its parent-relative transport delta is unchanged; 140 Python contracts, full Rust gates, and exact 100% local production function/line/region/branch coverage pass. Dispatch still proves neither the correlated result nor text-entry success, and fresh hosted exact-head checks plus parent-first integration remain required.
-- PR #271 is Draft at exact head `802ec806cdd4560eab48c484f435766ecabda353`, non-force synchronized onto #270 exact `191a14535219ea8033777fa4c970efb281b62418`. It admits only the exact typed observation response, treats unequal text as `PostconditionMismatch`, and retains no page-controlled or expected text in public evidence or diagnostics. Its test-first documentation contract, all 141 Python contracts, full Rust gates, and exact 100% local production coverage pass; hosted exact-head checks and ordered ancestor integration remain required.
-- PR #195 is Draft at exact head `48eb2d23009c1c804520dd5efcd0d4d072aacef1`. Its loopback regressions hold the accepted peer through opening-write timeout cleanup and locally revoked-stream classification, removing macOS close races without weakening production failures. The original affected test passed 50 consecutive focused regression passes, followed by 139 Python contracts, full Rust gates, and exact 100% local production coverage; fresh hosted exact-head checks and prerequisite integration remain required.
-- Opening-exchange fixture repair at #242 is Draft at exact head `2d0e9f69df9ade21d8e8e3d807c3ff644d83b310`, stacked on #195 exact `48eb2d23009c1c804520dd5efcd0d4d072aacef1`. A complete descendant Rust run reproduced a premature peer-close race in the mismatched-accept fixture after the existing invalid-deadline repair had been adopted. One shared test-only server now reads the complete request before replying and retains the peer through assertions in all three exchange fixtures. Fifty suite runs passed all 250 cases, followed by 139 Python contracts, complete Rust gates and exact 100% local coverage (809 functions, 7802 lines, 9959 regions, 942 branches). Production cleanup errors remain fail-closed. Native CI `33952463254` is queued; repeated local passes do not establish a zero flake rate or hosted acceptance.
-- PR #243 adopts that corrected parent by ordinary merge at `97fab641ed9d76e6c515eadcef0629edfc8064a3`, with exact base `2d0e9f69df9ade21d8e8e3d807c3ff644d83b310`. The parent-relative delta preserves the existing bounded frame transport and adds only its integration evidence; the shared fixture is identical to the owner version. All 139 Python contracts, full Rust gates and exact 100% local coverage pass (884 functions, 8733 lines, 11176 regions, 1008 branches). Native CI `33952588687` is queued, and the PR remains Draft. This repairs the previously stale #242-to-#243 dependency edge; later descendants still need ordered parent adoption and their own exact-head verification.
-- PR #246 adopts #243 at `585791f3641fbe757c3bd9fd36d5316adcc78d63`, with exact parent `97fab641ed9d76e6c515eadcef0629edfc8064a3`. Its ordinary merge preserves the message/JSON source and test blobs while carrying the owner fixture repairs. All 141 Python contracts, complete Rust gates and exact 100% pinned-nightly local coverage pass (962 functions, 9701 lines, 12427 regions, 1084 branches). Native CI `33953247053` is queued and this PR remains Draft. Fifty pre-integration fixture-suite runs passed, so no newly reproduced failure or measured failure-rate improvement is claimed. Raw protocol messages still carry no received-connection provenance or browser authority; child integration and hosted acceptance remain pending.
-- PR #247 exact head `6407895f4db4bee640074cb9c9d3cbe8b0e9e13a` merged as `b87191bcb6a95dfd7e0ed234e600639a1093c43a` into unprotected parent `feat/webdriver-bidi-text-message-assembly`; this is stack integration, not protected-main delivery. PR #248 is Draft at `b386f17c4826adabebda084bff2fba35aee94dd0`, now based on #246 `585791f3641fbe757c3bd9fd36d5316adcc78d63` by ordinary merge and explicit retargeting away from the merged #247 branch. Correlation production and Rust test blobs remain unchanged. Review reproduced that native unittest discovery collected zero checks from the free-function release contract; the same assertions now use the repository's TestCase format, collect one check and reject a missing release record. All 142 Python contracts, full Rust gates and exact 100% pinned-nightly local coverage pass (975 functions, 9848 lines, 12563 regions, 1092 branches). Native CI `33953719566` is queued. Previous Ready status and local totals did not prove enforcement of the uncollected check; current hosted verification, protected parents and received-connection authority remain unproven.
-- PR #249 is Draft at exact head `84b9407978ae0f6c115f01170b6069c601b21104`, ordinarily merged onto current #248 `b386f17c4826adabebda084bff2fba35aee94dd0`. The old `017d6e816f5a86544a63821b3ceaba94d5f17f44` tree lacked that parent and its newly discoverable release check; the parent adoption also reproduced a CHANGELOG-only merge conflict. The integration retains both release records, the parent's canonical opening fixture and native TestCase, and all four child-owned production/Rust-test blobs unchanged. All 142 Python contracts, the full Rust 1.97.1 gates and pinned-nightly 100% coverage pass locally: 989 functions, 9978 lines, 12717 regions, 1098 branches. Fresh exact-head CI `33954334610` is pending, not acceptance. The current #250 adoption is recorded next.
-- PR #250 now adopts #249 at `ec433b844a121f8554c062f92267991af9cacb6f`, with exact parent `84b9407978ae0f6c115f01170b6069c601b21104`. The conflict-free ordinary merge preserves all five child-owned production/Rust-test blobs and propagates the canonical fixture and native release TestCase. The pre-integration zero-test discovery RED becomes one executed check; all 142 Python contracts, full Rust 1.97.1 gates and pinned-nightly 100% coverage pass: 1023 functions, 10518 lines, 13525 regions, 1186 branches. Fresh exact-head CI `33955410724` is pending. Bounded status parsing still grants no authority and does not claim the later received-connection capability; current #251 adoption is recorded next.
-- PR #251 now adopts #250 at `f02af6d0dd01708d495cc08dec785675f3d58898`, ordinarily retaining both exact parents `86e8ad76838f2a64aa7e0cd56ba1f931c8d0c3dc` and `ec433b844a121f8554c062f92267991af9cacb6f`. Only the changelog required content reconciliation; both release records remain. The sender, public exports and both child-owned Rust integration tests are byte-identical to the predecessor. Native release-contract discovery first failed at zero and now executes one test; all 142 Python contracts, compileall, complete Rust 1.97.1 gates and enforced 100% coverage pass: 1037 functions, 10647 lines, 13681 regions, 1192 branches. The pinned branch-instrumentation warning below was independently reproduced on this tree too. Exact CI `33959168801` is queued; Draft, current-parent protection and hosted acceptance remain separate. A successful session-end frame write is not remote completion or process/profile teardown proof. Current #252 adoption is recorded next.
-- PR #252 now adopts #251 at `6569bf40b6595ac74c2f0a997d202137f07ba1db`, retaining predecessor `2015259529ada99af836989079cc85a15779a2d8` and current parent `f02af6d0...` by ordinary merge. The response implementation, public exports and four loopback response tests are unchanged; both conflicting release records remain. The inherited release-contract loader moves from zero RED to one executed GREEN, and all 142 Python contracts, compileall and complete Rust 1.97.1 gates pass. Enforced numerical coverage is 100%: 1044 functions, 10699 lines, 13751 regions, 1192 branches; the branch-instrumentation warning remains. Exact CI `33959789688` is queued. This response layer does not authenticate the received connection or prove resource teardown. #253's current-parent adoption is recorded next; the later #255 provenance repair remains separately owned.
-- PR #253 now adopts #252 at `afb623e4449b7cbf926fdcef7225ceaca822cfcf`, retaining predecessor `0d72082e595c0e1fcc03d609ba337896ed14e2fc` and current parent `6569bf40...` by ordinary merge. The assessment implementation, public exports and two child tests are unchanged; both conflicting release records remain. Native inherited release-contract discovery moves from zero RED to one executed GREEN. All 142 Python contracts, compileall and complete Rust 1.97.1 gates pass, with enforced numerical 100% coverage: 1054 functions, 10747 lines, 13787 regions, 1194 branches. The branch-instrumentation warning remains, and exact CI `33960670119` is queued. Review `5120013340` remains unresolved: caller-supplied teardown claims still cannot authenticate operational completion, regardless of local coverage. The #254 adoption below preserves #255's separately owned removal/provenance repair as a remaining prerequisite. This intermediate Draft is not security or operational acceptance.
-- PR #254 now adopts #253 at `b11b6c9bccd8335b58a4fb599f8ad29ac419637f`, retaining predecessor `cbaf50dcc97753cc73135497ea8225e8b18de190` by ordinary merge. The closure implementation, public exports and all six real-loopback child tests are byte-identical to that predecessor; both release records survive the changelog conflict. The inherited correlation release test moves from zero collected RED to one executed GREEN. All 142 Python contracts, compileall and complete Rust 1.97.1 gates pass, with numerical 100% coverage: 1061 functions, 10802 lines, 13843 regions, 1200 branches. Review `5120077272` remains actionable: closure kind/status still cannot establish that the acknowledged connection generation closed. Exact CI `33961562189` is queued; #255's current adoption below preserves its connection-provenance repair. Neither coverage nor loopback tests authenticate the missing connection binding, and the instrumentation warning remains.
-- PR #255 now adopts #254 at `3e7057443d7c9532ff526acb5eefe8cd4778c767`, retaining predecessor `ebac126d1632c94775c2454423575275eec45def` and parent `b11b6c9bccd8335b58a4fb599f8ad29ac419637f` by ordinary merge. All 15 owner production/test blobs are byte-identical to the predecessor. Native release-contract discovery moves from zero RED to one executed GREEN; 13 focused real-loopback tests, 142 Python contracts, compileall and the complete Rust 1.97.1 gates pass. Numerical 100% coverage is 1082 functions, 11032 lines, 14086 regions, 1202 branches. CI `33964381844` was queued at new-head observation. The previous unused-accessor repair is correctly attributed to `ebac126d...`, not its failing `63cbca0...` predecessor. Process/profile cleanup and hosted acceptance remain unproven.
-- PR #256 now adopts #255 at `ced4a851ca66c08d895a098725c7f0ad3ecf0c38`, preserving predecessor `9f2e6f29be46371762e3031a97c1cac04720694f` and current parent `3e705744...`. The core click implementation, exports and four tests are byte-identical; the correlation child delta remains only its typed click kind. Four click tests, 13 provenance/teardown tests, 142 Python contracts and complete Rust gates pass after inherited release discovery changes from zero RED to one GREEN. Numerical 100% coverage is 1088 functions, 11077 lines, 14146 regions, 1210 branches. CI `33965138401` and compatibility `33965138418` were queued; inert serialization remains distinct from authorized input or an observed browser click.
-- PR #257 now adopts #256 at `f4a8f2cbf515bea348f500b59615a9581c1b96a2`, preserving predecessor `ea2b5b78868917219c46f1304558b92490a7f6fe` and parent `ced4a851...`. The sender and two child-test files are byte-identical; the composition conflict retains both the sender and connection-bound reader. Ten focused tests, 142 Python contracts and full Rust gates pass after the inherited zero-to-one release-contract repair. Numerical 100% coverage is 1093 functions, 11131 lines, 14202 regions, 1214 branches. CI `33965440517` was queued. No focused retry was needed in this run; that does not repair the unchanged fixture's previously observed platform-level socket race. Frame-write completion is not browser-action completion.
-- PR #258 is Draft at exact head `5f830324f6d5a47ac213a57528ef95649bdfd0df`, retaining predecessor `f2ceabb3ea50b1959e936503c50cae12f3e6e480` and actual #257 `f4a8f2cb...` by ordinary merge. Its response implementation, four loopback tests and child-owned bounded socket-observation adjustment are byte-identical. Fourteen focused tests, 142 Python contracts and full Rust gates pass; numerical 100% coverage is 1100 functions, 11185 lines, 14272 regions, 1214 branches. CI `33965695405` was queued. The inherited release contract now executes. This click-response API remains protocol correlation only; the separate received-connection proof for session teardown does not automatically authenticate it or establish a browser post-condition.
-- PR #259 is Draft at exact head `66731982ed51ad62a04fcfbc759b0a33721a0254`, retaining predecessor `e1105ddf86f6c79443af8b4d306b9d34cb703c17` and current #258 `5f830324...`. Seven child production/test blobs remain byte-identical, including context admission, navigation projection and all three navigation suites. Fifteen focused tests, 142 Python contracts and full Rust gates pass after native release-contract discovery is restored. Numerical 100% coverage is 1140 functions, 11841 lines, 15129 regions, 1332 branches. CI `33965966736` and compatibility `33965966679` were queued. Exact context/URL matching is observation evidence, not transport authentication, click causation, origin rebinding or document advancement.
-- PR #260 is Draft at exact head `2c5049aff97a90958e8262b1d403bdcbd64a1e8b`, retaining predecessor `3a651967c421f77088fe25e86a63faae295390b3` and current #259 `66731982...`. The document-advance implementation, two loopback tests and Proposed ADR 0103 are byte-identical. Seventeen focused tests, 142 Python contracts and full Rust gates pass after the inherited release contract becomes executable. Numerical 100% coverage is 1154 functions, 11969 lines, 15314 regions, 1334 branches. CI `33966229736` was queued. Stale-epoch and retired-context rejection remain intact; registry advancement cannot authenticate the observation, bind the new origin or prove action causation. All six current integrations above retain the unstable branch-option warning, remain Draft and unmerged, and require their own hosted checks and policy-satisfied promotion.
-- PRs #250 through #257 remain Draft. The recorded fully locally verified chain used exact heads #250 `0eab23d5e388c5c8b984c0021a58316680c9ba8b`, #251 `86e8ad76838f2a64aa7e0cd56ba1f931c8d0c3dc`, #252 `2015259529ada99af836989079cc85a15779a2d8`, #253 `0d72082e595c0e1fcc03d609ba337896ed14e2fc`, #254 `cbaf50dcc97753cc73135497ea8225e8b18de190`, #255 `a13de5f9321e72c1867974eb7a43230f031e58df`, #256 `9f2e6f29be46371762e3031a97c1cac04720694f`, and #257 `ea2b5b78868917219c46f1304558b92490a7f6fe`. Each updated tree passed 141 Python contracts, the full Rust gates, and CI-equivalent pinned-nightly 100% production coverage. #250 and #257 each had one known macOS socket-observation race; their exact focused retries and complete coverage reruns passed. Exact-head review found that #253 still lets three unauthenticated caller booleans mint `OperationallyComplete`; raw teardown claims must remain explicitly unverified until session-bound transport, process-exit, and profile-removal evidence types are issued by their owning runtime boundaries. Hosted exact-head checks and ordered parent integration remain required. The current #255–#260 rows above supersede this older lineage only at their explicitly verified heads; no predecessor GREEN transfers to later hosted checks or protected-main delivery.
-- Reviews [#254](https://github.com/ContextualWisdomLab/OriginWeave/pull/254#pullrequestreview-5120077272) at `cbaf50dcc97753cc73135497ea8225e8b18de190` and [#255](https://github.com/ContextualWisdomLab/OriginWeave/pull/255#pullrequestreview-5120077213) at `a13de5f9321e72c1867974eb7a43230f031e58df` found a remaining connection-provenance gap. #254 passed six focused closure tests, formatting, strict network Clippy and network rustdoc locally. #255 removes the parent process/profile booleans and keeps overall teardown pending, but a new loopback regression fails when an acknowledgment from connection A is combined with closure from connection B: the assessment reports the acknowledged transport closed. Both connections can use command id 7; #254 retains only closure kind/status and the acknowledgment retains only its command id. Repair the established-stream, frame/message and command-correlation owners to preserve a non-forgeable connection-generation binding, carry it into acknowledgment and closure evidence, then reject mismatches in #255. Different-session and same-session reconnect regressions are required; endpoint equality, command-id equality and caller-supplied markers are insufficient. This is unresolved evidence integrity, not operational-completion authority or shipped behavior. Both PRs remain Draft with hosted checks queued.
-- Follow-up review `5120203295` at #255 head `92392b2da33bcd5446c6d2eb1b5504c39e60e3d6` established that received-message provenance must be checked before correlation is consumed. At that reviewed head, sender/closure generations rejected independently parsed evidence, but the two-real-socket response-substitution regression still failed: response B combined with outstanding-command state A became acknowledgment A because its generation came from the registry rather than the received text. The [exact-head review and runnable regression](https://github.com/ContextualWisdomLab/OriginWeave/pull/255#pullrequestreview-5120203295) preserve the RED lineage; the later repair evidence below supersedes this failure only for its explicitly tested snapshot.
-- Received-connection verification at #255 head `e7bfec4488b7cb4776df7b546cacb46c8c9eb13e` found that all 12 focused tests passed: the trusted reader retains one non-cloneable stream and assembler through fragments/control messages, the parser rejects a foreign generation before consuming the original command, and cross-connection closure cannot satisfy the acknowledgment. All 141 Python contracts and strict rustdoc passed. Full quality remained incomplete: formatting and Clippy failed, and pinned-nightly coverage rejected lines=11046/11049 and regions=14067/14071 even though functions and branches reached 100%. The first complete stable Rust run also exposed the inherited fixture race; its unchanged retry passed, which was not treated as repair. The fixture root cause is now repaired at #242 and integrated into #243 as recorded above. [Verification and exact uncovered paths](https://github.com/ContextualWisdomLab/OriginWeave/pull/255#issuecomment-5550208364) remain snapshot-specific; later #255 changes, hosted checks and descendant adoption need fresh verification. Overall teardown remains pending and none of this is protected-main delivery.
-- Follow-up quality verification at #255 head `63cbca0a98cf9496af981819d98029e656fc4342` completed locally at `2026-09-05T08:20Z`: 141 Python contracts, compileall and the complete Rust 1.97.1 tests passed. Formatting still failed, and strict Clippy rejected an unused private accessor left after the impossible provenance fallback was removed. Workspace check and strict rustdoc completed with that compiler warning. Pinned-nightly coverage enforcement failed at functions=1082/1083, lines=11046/11051, regions=14071/14074; branches=1202/1202. The missing-lines report identifies only the obsolete accessor's five lines; prior event/null-id and generation-exhaustion gaps are covered. The minimal owner correction is removal of that uncalled accessor plus actual pinned rustfmt, preserving received-connection validation and rejecting warning/coverage exclusions. All read-only validation handles are terminal and the existing #255 source owner has the diagnostics; no competing source change was made. Exact CI `33953476003` remains pending. These results supersede the earlier local snapshot only at this tested head, not at later repairs or protected main.
-- PR #139 at `b7ea5bfe336456fb263dd479a60b1cd0193d8a47`, based on #136 `1cffb2e23d4002f12e8462c4c8c24f404a4eeee7`, preserves all inherited documentation and limits its unique delta to the intended cleanup wrapper, regression and changelog. Both indexes retain ADRs 0007–0010 and ADR 0009 remains Proposed. All 179 Python contracts passed; [review evidence](https://github.com/ContextualWisdomLab/OriginWeave/pull/139#issuecomment-5550147096) records all three resolved findings. Current hosted Rust, coverage and pinned-Chromium runs were cancelled, so this Draft has no current hosted or browser-runtime GREEN from that verification.
-- PR #141 at `fbdf64f5818ce0c53b475196d5bcfa2ac9900846`, based on #139 `b7ea5bfe336456fb263dd479a60b1cd0193d8a47`, retains the reviewed failure/surface/timeout cleanup contracts. All 183 Python contracts passed; controlled probes confirmed that unexpected exceptions propagate after temporary-profile removal and cleanup-exit failures cannot return false cleanup evidence. The remaining two informational threads were resolved without source changes, leaving all four threads resolved; [review evidence](https://github.com/ContextualWisdomLab/OriginWeave/pull/141#issuecomment-5550364179) binds these results to the unchanged head. Hosted CI `33916583623` and pinned-Chromium run `33916583506` were cancelled. This is not current browser-runtime GREEN, adversarial erasure proof, counted approval or protected-main delivery; the PR remains Draft.
-- PR #142 at `015025f2539e4fb1dbd7d259ec22dad50f944396` passed fresh read-only failure-boundary review: 187 Python contracts passed, and controlled probes preserve unknown identity on observation failure, reject a still-live identity at the existing deadline, and accept absence or a different start-time as exit of the original identity. Both informational threads were resolved without production changes. Its exact-head CI `33916627099` and compatibility `33916627091` remain cancelled, not Linux/browser runtime GREEN.
-- PR #143 at `44fd9a450f864feff5cf2ba2883425a71ba10b9b` is a documentation and real-path regression repair on exact #142. The release-record check first failed for the missing failure-cleanup entry, then all five focused and 192 Python contracts, full Rust 1.97.1 gates and pinned-nightly 100% coverage passed: 415 functions, 3555 lines, 4444 regions, 476 branches. The production runner is unchanged. Observed exit and a surviving identity both retain failed-task status; process-observation errors leave termination unproven and record only the bounded observation-error type, losing the original browser-failure type in that fallback record. Private exception messages remain absent, and profile cleanup cannot turn a failed task into a pass. Both informational threads are resolved with this limitation documented. CI `33954715539` and compatibility `33954715568` are queued; controlled local probes do not prove real Linux/pinned-Chromium acceptance. Current #144 adoption is recorded below.
-- PR #144 at `09f2e087d0c20fe81386c18099e739a9e611a8ad` completed read-only informational review: 194 Python contracts passed, and controlled real-helper probes confirmed conservative failure when a descendant PID is reused before start-time capture, separate root and process-set deadline budgets, and the defensive exit-count invariant across all eight absent/live combinations of three descendants. Three informational threads are resolved with those limits recorded; no runtime code or timeout changed. CI `33916685294` and compatibility `33916685137` remain cancelled, not current browser acceptance. This is predecessor evidence superseded by the current adoption below; process identities appearing after sampling, cgroup ownership and OS-wide orphan absence remain outside this evidence.
-- Earlier #258 `f2ceabb3ea50b1959e936503c50cae12f3e6e480` and #259 `e1105ddf86f6c79443af8b4d306b9d34cb703c17` each passed 141 Python contracts and full Rust/numerical-coverage checks on the older #257 lineage. Their new current-parent integrations and independent 142-contract verification are recorded above; the historical measurements do not transfer.
-- PR #93 is Draft at exact head `0664f0452cb329cd692cce7f61f9001652abfda2`, based on #271 exact `802ec806cdd4560eab48c484f435766ecabda353` and synchronized with #242's complete opening-fixture cleanup. `SemanticNodeActionBinding` preserves semantic-node and business-origin identity, but does not authorize policy or execute input. Its 141 Python contracts, full Rust gates, and exact 100% local production coverage pass; hosted exact-head checks and ordered ancestor integration remain required.
-- PR #95 is Draft at exact head `97aa0f2e340ee6fd920d0418f97af276b190554f`, stacked on #93 exact `0664f0452cb329cd692cce7f61f9001652abfda2`. Only `Decision::Allow` creates policy-authorized value; other decisions fail closed, and registry-owned current authority returns `NotAdmitted` after document advance removes the node. The slice performs no browser I/O or postcondition proof. Its 142 Python contracts, full Rust gates, and exact 100% local production coverage pass; hosted exact-head checks and ordered parent integration remain required.
-- PR #96 is Draft at exact head `b7ba8dd1433410cee43084a73e31816da841b2a2`, stacked on #95 exact `97aa0f2e340ee6fd920d0418f97af276b190554f`. Registry-owned node authority is revalidated before one adapter callback, and the callback is never invoked after admission is removed. Adapter completion remains separate from postcondition proof. Its 143 Python contracts, full Rust gates, and exact 100% local production coverage pass; hosted exact-head checks and ordered parent integration remain required.
-- PR #101 is Draft at exact head `bc810b121bb0303f55afa8777a23cc0f9748c1db`, stacked on #96 exact `b7ba8dd1433410cee43084a73e31816da841b2a2`. A known-disabled interactive action fails as `NodeNotEnabled`, while `ScrollIntoView` remains selectable; retained observation state is neither current Chromium proof nor dispatch authority. Its 144 Python contracts, full Rust gates, and exact 100% local production coverage pass; hosted exact-head checks and ordered parent integration remain required.
-- PR #102 is Draft at exact head `a123c55d4839dae1db7e6671f7d4d158c7cfd9db`, stacked on #101 exact `bc810b121bb0303f55afa8777a23cc0f9748c1db`. Fresh semantic comparison rejects another node as `ObservationAuthorityMismatch`, removed action support, and newly disabled interactive state, but does not obtain or authenticate the observation or dispatch input. Its 145 Python contracts, full Rust gates, and exact 100% local production coverage pass; hosted exact-head checks and ordered parent integration remain required.
-- PR #103 is Draft at exact head `8b3416169346fa04b53c915b813d55ccf47d1876`, stacked on #102 exact `a123c55d4839dae1db7e6671f7d4d158c7cfd9db`. Same-call dispatch checks registry-owned browser authority first and fresh semantic state second; the callback is never invoked on either failure, and completion is not postcondition proof. Its 146 Python contracts, full Rust gates, and exact 100% local production coverage pass; hosted exact-head checks and ordered parent integration remain required.
-- Issue #201's release/SBOM lane remains dependency ordered, and GitHub Releases is empty. OriginWeave remains pre-GA until exact protected-head versioning, signed artifacts, SBOM/provenance, reproducibility, rollback, compatibility, security, and commercial-acceptance evidence exist.
-- The principal commercial gaps remain #27 Manifest V3/native-host isolation, #9 bounded HTTP/browser-network consumption, #10 purpose-bound protected-data runtime, #28 the first complete Chromium Agent Task vertical slice, #199 durable WARC/PROV and retention/replay, #200 stable BAP/MCP runtime API, #201 signed cross-platform distribution/update/rollback/SBOM/SLSA, #202 enterprise identity/tenant/policy/approval/audit/SLO operations, #203 exact-artifact commercial acceptance, #276 contextual-orchestrator migration, and #279 exact-head documentation verification without unnecessary Rust queue load.
-- Current evidence procedure remains fail-closed: queued reviewer evidence is non-passing; paginate the full PR inventory, bind checks/reviews/threads/workflow runs to each unchanged exact head and independently resolved live base, consult the active ruleset, and discard queued, skipped, cancelled, absent, predecessor, synthetic, status-only, and model-only evidence as passing proof.
-
-- Current PR #144 parent adoption is `3c0c5c363c2da0b4e8a621226e2f7766c735b743`, ordinarily integrating #143 `44fd9a450f864feff5cf2ba2883425a71ba10b9b`. The previous child collected three inherited failure-path contracts instead of five. Both release records and the parent's real failure-path regression now coexist with unchanged child production and process-set tests. All five focused and all 196 Python contracts, complete Rust 1.97.1 gates and pinned-nightly 100% coverage pass locally: 415 functions, 3555 lines, 4444 regions, 476 branches. Exact CI `33956094018` and compatibility `33956094011` are queued, not acceptance; the prior informational review limits remain unchanged.
-- PR #145 now adopts #144 at `bb5e8f834c37f9ce35f84db8ed1146da3659d6aa`. The ordinary merge preserves the child runner and both process-set/forced-close regressions while moving the missing inherited-contract count from three to five. All five focused and all 202 Python contracts, full Rust gates and the same four-dimension 100% coverage pass. Ten controlled actual browser-pass/trial probes distinguish observed root/set survival, observation errors and interrupted capture; session/driver cleanup alone does not prove process exit. Both informational threads are resolved with that corrected limit. Exact CI `33956596014` and compatibility `33956596013` remain queued.
-- PR #146 review repair is `812c0020bd2ecdb3eda39d999ed3327647550bdf`. Five new tests first reproduced 10 assertion failures and 7 uncaught protocol-error cases despite the predecessor's 216 passing contracts. The repair restores the non-boolean pre-shutdown count range, preserves only validated known ordinary cleanup fields, records typed terminal protocol errors without remote messages, and removes the obsolete HTTP-body close recognizer. Startup retry policy, request redaction and owned-process deadlines remain unchanged. Six new behavioral tests and all 224 Python contracts, complete Rust gates and the same 100% production coverage pass. The concurrent parent-adoption `11944410450684809ee1a71a35c77abafc5358db` is retained through an ordinary merge whose tree is identical to verified repair `d636a1829332610ada458df7b5b9d267f038a2f8`. All four review threads are resolved; exact CI `33957036553` and compatibility `33957036650` are queued, not browser acceptance. #147 must adopt this parent while retaining its own shared-deadline behavior. These three PRs remain Draft and no protected delivery, eligible approval or release is claimed.
-- Follow-up #287 review evidence at `2026-09-05T09:07Z`: Strix scan job `101243872506` executed successfully, including its quick scan and report upload. Noema job `101245089068` failed with HTTP 502 after 2739.2 seconds and one caller attempt. A fresh exact-head inventory showed no duplicate Noema run, so the same failed job was queued for a supported rerun; attempt 2 now has successful exact-head admission and queued review job `101282290732`. Queue admission is not provider recovery, executed review, a verdict or eligible approval. No caller timeout, provider, model, workflow or quality gate changed. The three central CodeQL dispatches remain queued and were not duplicated. This supersedes the earlier in-progress Noema/Strix snapshot above without transferring any old result to a new source head.
-
-- Prior #255 source-owner repair is `ebac126d1632c94775c2454423575275eec45def`, verified at that earlier owner snapshot before current-parent adoption. Its 11-file delta removes the obsolete private accessor, applies pinned formatting and repairs the Rust documentation link without removing received-connection validation. The source owner reports 141 Python contracts, full Rust 1.97.1 verification and coverage functions=1082/1082, lines=11025/11025, regions=14071/14071, branches=1202/1202. This is the owner's historical exact-tree local verification, not an independent rerun by this documentation lane or a transfer of earlier GREEN. At that observation, hosted CI `33957423557` was queued; the earlier `63cbca0...` quality RED is retained as historical RCA, and the new independent parent-adoption results are recorded above.
-- The pinned cargo-llvm-cov 0.8.6 branch report emits `warning: --branch option is unstable`. The warning was independently reproduced when reporting the existing #144/#145/#146 measurements, whose numerical verifiers still pass. Thus numerical coverage enforcement is distinct from warning-free instrumentation. The #255 source owner likewise leaves warning-free measurement acceptance outstanding. No warning suppression, coverage exclusion, denominator manipulation or alternate unreviewed measurement is used; compiler/Clippy/rustdoc results, instrumentation maturity, hosted acceptance and release readiness remain separate.
-- Central Strix #1563 advanced to `13fbb48e0b3eeca4ce7d9678add934f9bd87ad3f` from `1221b1604e1a6cfde8ca5ab7fd3e93e0fe9faf69`. The independently inspected four-file source diff adds a completed-artifact regression using the exact #166 denied-report prose and restricts console denial matching to explicit diagnostic lines while retaining warning, fatal, and workflow-error signals. This establishes a source repair, not an executed Origin acceptance result. The focused and full shell harness results plus 2890 passed, 1 skipped, and 21 subtests are owner-reported; this consumer lane did not execute them, and the skipped case is not passing evidence. At 13:46 UTC, ordinary merge `eaf9594f7fe8d8e1994349289183d6cbad056579` retained `13fbb48e...` and adopted protected central main `6f8c51d7389c22ebaf294fe8fe9ef495257883c0`. That fresh head is unmerged and hosted checks are queued; its cancelled predecessor checks and earlier local results do not validate the new merge. No Origin #166 consumer rerun was performed, so its historical executed failure remains unreplaced. Actual #219/#240 gateway and hosted execution-limit failures remain distinct.
-- Historical #147 `46e75859ef5aff2054fd6ca1720d518438e97d0f` included #146 `812c0020...` in ancestry but dropped the parent's six review-evidence regressions and reversed their associated production repairs. Native discovery collected zero tests; read-only replay of the exact parent module against that child reproduced 10 assertion failures and 8 uncaught protocol-error cases. The merge title, ancestor presence and earlier 222 passing child tests did not establish successful inheritance.
-- At predecessor #147 `af1b98ba377c73b88baa9633e2232e7f76e76f36`, the external synthesis restored parent runtime repairs and all six methods while preserving the combined forced-close observer. Its recorded RED is that all 228 native tests execute but two forced-close protocol-fault subcases fail. The inherited test injected the individual root waiter although the forced-close lane called the combined observer; a fresh focused reproduction took 10.417 seconds. The [review evidence](https://github.com/ContextualWisdomLab/OriginWeave/pull/147#issuecomment-5550841022) retains this failure and its subsequent repair rather than treating recovered ancestry as acceptance.
-- Current #147 `3dff28d9bf2dd27b72507e39979d51b8bf140fb4` repairs only that lane-specific test injection and its documentation. Ordinary cases retain the individual False result; forced-close cases inject the combined (False, False) result and assert the exact root-only identity tuple. Every existing assertion and all six methods remain; all six focused and all 228 native tests now pass, along with compileall, complete Rust 1.97.1 gates and the same 415/3555/4444/476 numerical 100% coverage. Production, deadlines, retry policy and failure denominators are unchanged; the branch warning remains explicit. Writer reassessment occurred after the temporary route closed and contacted tasks confirmed read-only status; the current repair is a normal forward push. CI `33959982049` is pending and compatibility `33959982033` is queued. The ordinary-pass two-deadline finding stays unresolved in the separate #150 scope; no hosted, protected-main or real-browser acceptance is claimed.
-- Current #148 `bded4fc9d32e5e047cea99d182aa05ae2cd03bf6` adopts that exact #147 by ordinary merge and repairs review `5120692604`: a real crash-trial startup probe first failed because the emitted browser arguments disabled Chromium's sandbox. Removing that single argument preserves the startup-failure path, driver termination and temporary-profile removal, with no unsandboxed fallback. The inherited six review regressions and both child crash-cleanup/exit-detection test blobs remain intact. The 247-test macOS run has 244 passes and three explicitly skipped Linux pidfd cases; all 247 Python contracts pass on Linux with zero skips, including a killed unreaped child, non-terminating signal and stale process identity. Complete Rust 1.97.1 gates and numerical 100% coverage pass at 415/3555/4444/476, with the branch warning retained. CI `33962062616` and compatibility `33962062608` are queued. Three other launch lanes still disable the sandbox in this branch and require their separate owner repairs; this is not runner-wide sandbox completion, pinned-Chromium acceptance, review dismissal or protected delivery.
-- Current #150 `d3c29359fd10e540ca9b3b2723bde6f94866cdb6` adopts the same #147 while preserving its existing combined ordinary-pass observer and both child test blobs. The inherited six review methods were absent before adoption. Their restoration exposed two ordinary protocol-fault failures, and full replay then exposed three sibling failure-observation cases still injecting the superseded individual waiter. Test-only reconciliation now injects paired observer outcomes and asserts the exact root-only identity tuple; incomplete process capture cannot claim full-set termination. No production timeout, retry, polling rule, success condition or denominator changed. All inherited and child checks execute, and all 229 Python contracts pass on macOS and Linux. Complete Rust gates and the same four-dimension numerical 100% coverage pass with the instrumentation warning retained. CI `33962429156` and compatibility `33962429172` are queued. The ordinary-pass deadline finding remains valid on #147 until this separate child is actually integrated; it was not resolved early.
-- Supplemental Linux verification used the existing Colima guest (Python 3.12.3, kernel 6.8.0-117-generic), without installing dependencies or changing VM configuration. A first macOS archive introduced binary AppleDouble files and caused two unrelated TLS source-decoding failures; exporting the same tracked files without filesystem metadata repaired the transport, not the tests. Final host/guest tracked-file SHA-256 manifests match exactly: #148 `c808905cbea16fd85b9519fa0a7c431d032f6692bea172950e5ba23613e7e98b`, #150 `0f5f93ed8789843ca4f228d1117328d029e3fab461f0450c75964f2ef6b98324`. These are actual Linux contract runs, not actual pinned-Chromium browser trials, hosted CI or release evidence.
-- #166 Strix run `33929688857` / job `101237371800` failed at `2026-09-05T10:53:36Z` on unchanged `e84a1a2...`. Its executed trusted central source was `a9aeee8fc94ad6002a059b380b268590ce496ef0`. The final scanner attempt exited zero, recorded completion from `10:25:17Z` to `10:53:29Z` and emitted zero SARIF results, but the console classifier also matches ordinary denied-policy prose. Replaying its exact first predicate against the sanitized final console log matches two policy-description lines and independently explains the final provider-unavailable classification. The final scanner report log has no corresponding failure-token match; the initial report-directory hypothesis was rejected. Artifact `9968177796` has SHA-256 `1ed63eb46e83d8acb61e281a6563400269a45c0924ae989ee30a4addc041048b`. Earlier attempts did have errors, so neither all-attempt success nor authoritative security acceptance is inferred. The reproduced classifier defect and exact evidence were sent to the existing central owner; no central source, provider, timeout, gate or required status was changed here, and repeating the same classifier was not used as a remedy.
-- #219 Strix run `33925442296` / job `101226800608` on `65e4315...` is a distinct failure: a real gateway HTTP 500 occurred at `08:08Z`, and the job was cancelled at `10:51:35Z` with the six-hour GitHub job execution limit explicitly recorded in its check annotation. This is neither a passing scan nor evidence that the #166 console defect caused that cancellation. Its already admitted Noema attempt 2 remains a separate queued item. Canonical provider/orchestration and runner feasibility must be reassessed before another attempt; no caller timeout or model fallback was expanded.
-- #240 Strix run `33925596890` / job `101227893828` on unchanged `24930a3...` was also cancelled at `11:20:05Z`; its annotation independently confirms the same six-hour hosted execution limit. Its actual console records gateway HTTP 500 at `07:01Z` and `10:34Z`, followed by cancellation at `11:19:59Z`. These are observed failures, not a generic inference from a red status. No successor Strix run existed in the refreshed exact-head inventory, but an unchanged retry would not repair the canonical provider or classification defects; this source/documentation lane did not start one or alter execution limits.
-
-- Temporary integration PR #291 is now closed after an external merge into `automation/147-merge-base-20260905`, whose base was #146 `812c0020...`. Live API metadata records head `52575632de07bcb90791e491b9a64b6875532ebe` and merge `24fad0498a49267982805d35d2dd57484efb4671`; Git confirms that merge is an ancestor of current #147 through forward integration `07369a37b54c11c32f810c3d5c107ad6d4b9feb4`. The temporary route neither replaced #147 nor reached protected main. The recovered runtime and test content, subsequent test-reconciliation repair and ordered acceptance requirements are evaluated separately from its merged/closed label. This lane performed no merge or closure. The fresh full inventory is 125 open / 12 Ready / 113 Draft.
-
-## Observed snapshot: 2026-08-29
-
-### Protected-main truth
-
-- Protected `main` is at `542ca1e9c0a863595b8b6697790005d2471f5413` for this snapshot. Since the 2026-08-26 observation (`b05d5acca82b9d916ada2c8e82f59f92a89817e1`), protected `main` absorbed #161 (TLS trust-bundle identifier shape). PR #170's conservative `tools/list` discovery contract is also merged into protected `main`; the complete MCP adapter remains planned.
-- Phase 0 remains complete as a reusable safety-kernel foundation: typed policy contracts, destination classification, direct TCP peer verification, TLS service identity, evidence bounds, resource mitigation, document-node authority, and protected-main tests.
-- Phase 1 is **in progress**, not shipped. The first real Chromium vertical slice still needs the active WebDriver BiDi transport stack to reach protected `main`, then compose isolated Chromium launch, session/context identity, semantic observation, typed action authorization, native browser input, post-condition proof, evidence, cancellation, crash recovery, and profile/process teardown.
-- HTTP/1.1 bounds, downloads/MIME, proxy/PAC consumption, full browser-network integration, the sensitive-data broker runtime, durable WARC/PROV capture, persistent task/API surfaces, signed cross-platform distribution, enterprise administration, and release-grade buyer acceptance remain open.
-- Active pull requests remain evidence, not shipped behavior. Successful checks on a feature or stacked branch do not prove that protected `main` contains the capability or that a child can merge before its prerequisite.
+Protected-main truth is distinct from active implementation. The reviewer-provisioning gap remains an acceptance concern whenever no eligible non-author counted approval is present on a candidate; bot or advisory review text does not silently satisfy the ruleset.
 
 ### Open pull requests
 
-The live repository contained **108 open pull requests: 24 non-draft and 84 draft** when this snapshot re-paginated the complete open inventory. Compared with the prior **2026-08-28 111-PR snapshot**, stacked evidence PR #53 was merged into its unprotected feature parent; the preceding **2026-08-28 116-PR snapshot** had already recorded #71, #154, #233, #234, and #235 merging into their unprotected feature parents. PR #217 was squash-merged into the unprotected #210 feature parent, followed by PR #67 into the unprotected #64 feature parent; the current queue therefore contains no protected-main shipment. These counts are queue evidence, not protected-main delivery; protected `main` remains `542ca1e9c0a863595b8b6697790005d2471f5413`, with 11 open issues and no releases or tags. The volume and stack depth remain themselves a product-delivery risk: review, exact-head checks, dependency order, and integration truth can drift faster than a buyer-visible vertical slice reaches protected `main`.
-
-#### 2026-08-29 maintenance-loop record
-
-This snapshot re-fetched the complete open-PR inventory, the protected `main` commit, the active required-workflow ruleset, collaborator permissions, and the exact base/head pair for each representative PR. Before PR #238 was published, the same maintenance loop observed 115 open pull requests (31 ready, 84 draft); a later 116-PR recheck preceded five stack merges, leaving 111 open pull requests (27 ready, 84 draft), and subsequent child-stack merges left 108 open pull requests (24 ready, 84 draft). The active ruleset still requires one counted approving review and resolved threads; the only repository collaborator is `seonghobae`, so review provisioning remains the merge blocker for main-targeting PRs. PR #170 is merged and is not active-PR evidence.
-
-During this recheck, #71, #154, #233, #234, and #235 were merged only into their unprotected feature-parent branches after exact current-head checks and review-thread resolution. Their successful stack checks are not protected-main delivery, and their child branches retain independent evidence requirements.
-
-The sensitive-data child slice #53 was subsequently squash-merged into the unprotected #46 feature branch at merge commit `93c713a107df05385f745db4dca20091f21c4a3a`; #53 was exact head `4ecc81e59ae7bc3a640e65e2442bf30c079bd94c`. PR #46 remains an active main-targeting parent. Its current exact base/head pair is protected `main` `542ca1e9c0a863595b8b6697790005d2471f5413` to `373113119446d99f578febd39efc19366e7736b1`; the head adds the ADR 0007 predicate-boundary clarification and regression contract, with local Python/Rust verification green. Current hosted evidence remains incomplete: automatic OpenCode run `33189822385` / job `98913006386` failed closed without a current-head verdict, and central Strix run `33190794267` / job `98915422837` failed closed after three provider HTTP 500 attempts without a vulnerability report. A direct central `opencode-review` dispatch run `33192478312` / job `98921183278` was rejected because repository_dispatch actor `seonghobae` did not match configured scheduler identity `github-actions[bot]`; no qualifying non-author approval is present.
-
-The WARC/PROV child slice #217 was squash-merged into the unprotected #210 feature branch at merge commit `66f360ccac5cec60c72222cc79d58e39f6f00088`; #217 was exact head `6b8a3fdeae52ad94b90086bbc9b42863b90c9614`. PR #210 subsequently advanced to current exact head `7946dce9a3dd074047d93fca299d48c7aef40e47` after its merged-child attribution repair, recursively encoded-control repair, and exact coverage repair; this stack remains active-PR evidence and not protected-main delivery or approval evidence.
-
-The browser-task interruption child slice #67 was squash-merged into the unprotected #64 feature branch at merge commit `5021d142583cb5a8e393248048bb824762a98056` from exact PR head `25ab76e8279d4a904d04afeb264bac3e89f46b45`. PR #64 consequently advanced from `debc761aa59aee1509b7a260474fa33216453511` to current exact head `5021d142583cb5a8e393248048bb824762a98056`; its exact-head hosted checks were regenerating at this snapshot, with no unresolved inline review threads. This stack remains active-PR evidence and not protected-main delivery or approval evidence.
-
-The later exact-head recheck also recorded the WARC/PROV retention-lifecycle slice: #239 is Draft at `e840ca299d29a15223c8b9bb1397002c4f41b4a3` on #227 head `e45cd6cdcdee73b5c16dc942e6c98cb7e745fae0`, and #237 is Draft at `2459af602e72fbfe1ce816919473a1075ec0c41f` on protected `main`; their current exact checks and reviews remain independently actionable evidence. Neither active branch is protected-main behavior.
-
-The same exact-head recheck corrected workflow provenance: the active ruleset's seven required workflow entries (`close-empty-pr`, `opencode-review`, `pr-review-merge-scheduler`, `security-scan`, `strix`, `sast-semgrep`, and `noema-review`) point to the central `.github` repository, repository ID `1274066402`, where all seven files exist on `main`; their absence from the OriginWeave tree did not prove missing workflow identities. On PR #210 head `0341079331f9cea669eb9a5cc21842fd6027431e`, run `33177641855` failed closed because no OpenCode current-head verdict existed, while run `33177641888` failed closed after three bounded attempts because the Strix provider/backend returned an internal server error and produced no vulnerability report. A `gh run view` workflow-endpoint 404 for the external workflow IDs is a lookup mismatch, not passing evidence or proof that the identities are absent. A bounded central scheduler dispatch was accepted as run `33178984025` with auto-merge and branch updates disabled; no result is transferred until the exact head is revalidated. These failures block affected PRs until current-head review and security evidence complete. This does not authorize bypass, self-approval, stale checks, or weaker gates.
-
-The WARC resource-record slice #210 is now Ready; PR #210 current exact head is `7946dce9a3dd074047d93fca299d48c7aef40e47` based directly on protected `main` `542ca1e9c0a863595b8b6697790005d2471f5413`, after incorporating #217, correcting the merged-child attribution, and closing the recursively encoded-query-control coverage gap. Its predecessor head `5f59947f5e4b0d3bc0aa5b2d4c6722d3b7c43047`, prior stack merge head `66f360ccac5cec60c72222cc79d58e39f6f00088`, earlier exact head `bea65643109449d63d367a35b8d9bf327ee7cb2c`, and their OpenCode/Strix provider failures remain historical evidence only. At the current head, `Rust contracts` job `98942518975` and `Production coverage` job `98942518680` succeeded, while `noema-review` job `98942513421` and `strix` job `98942803402` remain in progress; the current-head OpenCode verdict is still absent and no counted approval exists. Central repair PR #1391 was opened at historical head `e4ba6b599cd1e50d0139762885682607b731655d` and is now open at exact head `36ac3aa71b2580685f84d416a81e42c39dee927c` on current central `main` `e1b03eebc6dc5c85aed393e5928927c96376cf46`. Its branch-update merge and follow-up prompt hardening are not approval or coverage evidence.
-
-The documentation refresh PR #238 remains Ready on protected `main` `542ca1e9c0a863595b8b6697790005d2471f5413`; its moving current head is intentionally authoritative in live GitHub metadata and the PR body rather than repeated in this self-referential baseline, while local full-suite evidence is green. The immediately preceding PR #238 head `d0b0d1ed92f891f14646fc673b8e1c0d912586fd` remains historical: automatic OpenCode run `33193822920` / job `98926243116` failed closed without a current-head verdict, current Strix run `33193822929` / job `98925769697` succeeded, and central dispatch run `33194506918` / job `98928580387` also failed closed at OpenCode after its validator, bootstrap, and coverage jobs succeeded. The preceding exact-head OpenCode run `33184553025` / job `98894986761` and earlier targeted run `33182749298` remain historical; targeted run `33182749298` dispatched OpenCode run `33182772296`, which failed closed as `MODEL_OUTPUT_UNAVAILABLE` with `model pool exhausted`, and optional cross-repository status publication was denied with HTTP 403. The current Devin documentation-evidence finding requiring an enforcing `tools/list` traceability contract has been implemented and its review thread resolved. No non-author counted approval exists, so these are review-tool/documentation findings, not approval or protected-main shipping evidence.
-
-The controlled Chromium prerequisite stack was also re-fetched after its exact-head repairs: #70 is Ready at `441a8ce1d09c329c5c1168f4906d9a38fd0abc01` on protected `main` `542ca1e9c0a863595b8b6697790005d2471f5413`; #71 is merged into the unprotected #70 feature branch; #72 is Draft at `600d3975c02b68da1974a4c73069b966b39dce7b` on the retained #71 branch; and #73 is Draft at `ce1b138509ab4f52cb0f80290f104358473c6ed3` on #72. #70's Rust, coverage, pinned-Chrome, and ordinary security checks are successful, while exact-head OpenCode failed closed without a current verdict and exact-head Strix failed closed after three provider HTTP 500 attempts; all current Devin informational threads are resolved. #72 and #73 retain independent exact-head evidence requirements. #82 is Ready at `f5776f5f233ac0a7c05e3f4a2846436c23438043` on protected `main`; its Rust, coverage, Chrome, and ordinary security checks pass, exact-head OpenCode failed closed without a current verdict, and its current Devin informational thread is resolved. #152 is Ready at `81407a0e5189a413d1be0963fea90a0c2f254ce1` on protected `main`; its source, coverage, and security checks are successful except exact-head `opencode-review`, which failed closed without a current-head verdict. These are active-stack evidence only, not protected-main behavior or merge authorization.
-
-#### Current exact-head active PR evidence
-
-The following representative slices were re-fetched from GitHub for this snapshot. Their exact base/head pairs are recorded so later checks, reviews, and restacks cannot be confused with predecessor evidence:
-
-| PR | State | Exact base head | Exact head |
-|---|---|---|---|
-| #73 | Draft | `600d3975c02b68da1974a4c73069b966b39dce7b` | `ce1b138509ab4f52cb0f80290f104358473c6ed3` |
-| #72 | Draft | `f86ce504138e79d6e95141a441f60b40920e1fa6` | `600d3975c02b68da1974a4c73069b966b39dce7b` |
-| #46 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `373113119446d99f578febd39efc19366e7736b1` |
-| #70 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `441a8ce1d09c329c5c1168f4906d9a38fd0abc01` |
-| #82 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `f5776f5f233ac0a7c05e3f4a2846436c23438043` |
-| #210 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `7946dce9a3dd074047d93fca299d48c7aef40e47` |
-| #64 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `5021d142583cb5a8e393248048bb824762a98056` |
-| #237 | Draft | `542ca1e9c0a863595b8b6697790005d2471f5413` | `2459af602e72fbfe1ce816919473a1075ec0c41f` |
-| #239 | Draft | `e45cd6cdcdee73b5c16dc942e6c98cb7e745fae0` | `e840ca299d29a15223c8b9bb1397002c4f41b4a3` |
-| #229 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `0145ccba5901e301b41d4be674ca1ed23483ad37` |
-| #220 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `b11db2be68f9b6d71aa4c4290b97a8b22097b353` |
-| #211 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `52a918577958a5701e1146c7eb8b62fe8f8ccd44` |
-| #152 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `81407a0e5189a413d1be0963fea90a0c2f254ce1` |
-| #195 | Draft | `6922dd98779e8f8aad132a3b1f563d7ba6e6d070` | `48eb2d23009c1c804520dd5efcd0d4d072aacef1` |
-| #242 | Draft | `48eb2d23009c1c804520dd5efcd0d4d072aacef1` | `55fef0c3fae1724eddada53e52c4a0311f509aa3` |
-| #248 | Ready | `6407895f4db4bee640074cb9c9d3cbe8b0e9e13a` | `7d6db16b2ead201fcec320854923f90d3ad0d8bc` |
-| #249 | Draft | `de7754aaeb97ccb0fd47bcbe1c4d99c10eaf84eb` | `2279d18189fcd6cdb2b38aca53b877434d41c913` |
-| #250 | Draft | `2279d18189fcd6cdb2b38aca53b877434d41c913` | `cbddf507ac41080ee65230a8d6047dd8d06fd719` |
-| #251 | Draft | `cbddf507ac41080ee65230a8d6047dd8d06fd719` | `99e2eb946e7ebbffa68f65a00d25243a2cf4242a` |
-| #252 | Draft | `99e2eb946e7ebbffa68f65a00d25243a2cf4242a` | `881a599fb3a920b6bfd4f0a276f3cf24a61d8194` |
-| #253 | Draft | `881a599fb3a920b6bfd4f0a276f3cf24a61d8194` | `5e2b17c7a8d1953bb06a41e0296f801f0d015a9a` |
-| #254 | Draft | `5e2b17c7a8d1953bb06a41e0296f801f0d015a9a` | `a4841016b94cb18e917d250a9ca9149af54ceef0` |
-| #255 | Draft | `a4841016b94cb18e917d250a9ca9149af54ceef0` | `f8edec38cf8ab7fde22b8d1de9305728c1a2f25b` |
-| #256 | Draft | `f8edec38cf8ab7fde22b8d1de9305728c1a2f25b` | `bd1f5ac60a76d2edb35e63095d406e53bc43931f` |
-| #257 | Draft | `bd1f5ac60a76d2edb35e63095d406e53bc43931f` | `ac73abfe7edd5786a7eb3eaab1a8c773093be7d3` |
-| #93 | Draft | `802ec806cdd4560eab48c484f435766ecabda353` | `0664f0452cb329cd692cce7f61f9001652abfda2` |
-| #95 | Draft | `0664f0452cb329cd692cce7f61f9001652abfda2` | `97aa0f2e340ee6fd920d0418f97af276b190554f` |
-| #96 | Draft | `97aa0f2e340ee6fd920d0418f97af276b190554f` | `b7ba8dd1433410cee43084a73e31816da841b2a2` |
-| #101 | Draft | `b7ba8dd1433410cee43084a73e31816da841b2a2` | `bc810b121bb0303f55afa8777a23cc0f9748c1db` |
-| #102 | Draft | `bc810b121bb0303f55afa8777a23cc0f9748c1db` | `a123c55d4839dae1db7e6671f7d4d158c7cfd9db` |
-| #103 | Draft | `a123c55d4839dae1db7e6671f7d4d158c7cfd9db` | `8b3416169346fa04b53c915b813d55ccf47d1876` |
-| #124 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `fdb88698ca20626a6643bc2ad7944fb968835700` |
-| #37 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `5e3dfcbd7a4daea297782cb99635990368589232` |
-
-These rows are delivery evidence only. None has counted independent approval in the current collaborator inventory.
-
-The current documentation branch is PR #238 itself, so its self-referential exact-head row is intentionally omitted; GitHub PR metadata and the PR body are the authoritative current-head record for this change.
-
-#### Historical 2026-08-26 maintenance-loop record
-
-The interactive maintenance loop performed the following verified state changes on exact heads; none of them is protected-main behavior until merged:
-
-| Action | Exact evidence |
-|---|---|
-| Supersession closure | #153 closed with replacement evidence: base-stack tip (`4da223ac`) already implements `_terminate_owned_process_bounded` exit-race tolerance that supersedes the branch delta |
-| Conflict reconciliation | Merge commits pushed to #37 (`27f6acd6`, ci.yml aligned to reviewed `nightly-2026-08-18` pin), #149 (`7852a540` + rustfmt fix `54f96008`), #152 (`65b0c705`), #173 (`ecc9574a`), #175 (`765c88f6`, keeps `crate_root.rs` naming) |
-| Governance remediation (#212) | #43 reconciled with main in `04e262d5`; the `chrome_sandbox` workflow mutation was first removed, then restored under recorded independent authorization (issue #212 option (b)) because the PR's own contract test fails closed without it; fresh exact-head checks re-ran on the restored head |
-| Security finding fix (#124) | Strix vuln-0001 (Unicode homoglyph path confusion, MEDIUM) remediated in `30cc458b`: audited workflow paths now restricted to a canonical ASCII alphabet with homoglyph/fraction-slash/fullwidth regression contract tests; CHANGELOG updated |
-| Fail-closed provider re-dispatch | ~21 failed Strix required-check runs re-dispatched on unchanged exact heads; completed reruns returned success on #46, #48, #156, #157, #159, #218, and #219 heads at snapshot time; cancellations only where newer heads superseded the run |
-| Current-head review re-dispatch | Central merge-scheduler dispatches sent for #47, #62, #63, #65, #74, #166, #173, #175, and #220 because their stale `CHANGES_REQUESTED` verdicts cited coverage-evidence results that are green on the same heads today |
-
-#### Organization review-pipeline congestion record
-
-Between 2026-08-26T02:44Z and 2026-08-26T03:35Z the organization-wide Actions queue exhibited a systemic backlog: scheduler, OpenCode-review-dispatch, Noema, and Strix runs across `.github`, `naruon`, `pg-erd-cloud`, and OriginWeave sat `queued`/`pending` while only single-digit runs were `in_progress`. This delays every current-head AI review and therefore every ruleset-gated merge. It is an infrastructure-capacity signal, not a code defect, and it does not authorize merging without current-head review evidence.
-
-Representative active workstreams at this snapshot were:
-
-| Workstream | Representative active PR evidence | Delivery boundary |
-|---|---|---|
-| Product baseline | (merged: #196 on 2026-08-24) | Baseline publication reached protected `main`; this document is its successor snapshot |
-| Presentation identity | #229 at `fb868589d065c2cea0b9c8c0f5e655a89f42bee6` onto `542ca1e9c0a863595b8b6697790005d2471f5413` | Ready/non-draft local privacy kernel; current required checks include a failed Strix run, and the PR remains blocked without counted approval; no protected-main shipment is claimed |
-| Enterprise approval authority | #220 at `b11db2be68f9b6d71aa4c4290b97a8b22097b353` onto protected `main` `542ca1e9c0a863595b8b6697790005d2471f5413` | Ready/non-draft bounded maker-checker approval lifecycle on the exact `ApprovalScope`; current checks and review state require a fresh exact-head audit, with no counted approval |
-| Release artifact identity | #218 and #219 | Ready/non-draft fail-closed benchmark release decision and canonical release manifest binding; Strix provider-failure reruns completed green on both heads |
-| Schema-bound extraction and BAP lifecycle | #209 and #208 | Ready/non-draft schema-bound extraction contract and resumable task-lifecycle kernel; #209 Strix rerun green, #208 rerun re-dispatched after a further provider failure |
-| WebDriver BiDi transport | #188 through #205 | Active stack whose top #205 merged into its prerequisite branch, not protected `main`; it exercises framed `locateNodes` exchange over a bounded WebSocket opening path, but authenticated browser-process provenance, semantic task execution, and protected-main shipment remain unproven |
-| MCP adapter | (#168 and #170 merged) | Typed MCP routing and conservative `tools/list` cache metadata are protected-main behavior; the complete MCP transport, OAuth, browser I/O, and persistence adapter remains planned |
-| Workflow-registry audit | #124 | Real Strix finding vuln-0001 (Unicode homoglyph path confusion, MEDIUM) remediated on head `30cc458b` with regression contract tests; fresh exact-head checks and review re-running |
-| Controlled Chromium and recovery | #65, #70, #72-#73, #100, #105, #142-#152 and descendants | Real pinned-browser fixture, semantic location, resource, crash, and teardown evidence exists on active stacks; #71 is merged only into the #70 feature branch, and evidence does not transfer across heads or prerequisites |
-| Durable WARC/PROV evidence | #210, #217, #239 | Bounded WARC resource records, PROV JSON-LD binding, and retention-lifecycle boundaries are active-PR foundations; durable ownership, replay, retention/deletion, and browser side-effect reconciliation remain open |
-| Manifest V3 and native messaging | #27, #43 governance remediation, and the extension/native-host stack including merged #154 and active #169 | Compatibility and Agent-authority isolation remain incomplete until exact release artifacts and platform matrices are proven; #43's sandbox workflow mutation is now owner-authorized under issue #212 option (b) |
-| Sensitive-data and model route policy | #10 and its active policy stacks | Deterministic policy values exist, but trusted broker execution, retention/deletion, runtime isolation, and auditable product workflows remain open |
-| VPN/profile intent | #149 | Bounded WireGuard/IKEv2 profile authority reconciled with main (`54f96008`); it does not create a tunnel, route, DNS state, authenticated gateway, or connectivity proof |
-
-PR #205 head `f427aa69151987d7e3369bd96d5739ea38d0f7ad` merged as `6c5ef5e2079d54c617183ecfa757e406f48f0aea` into stacked prerequisite branch `feat/webdriver-bidi-websocket-frame-transport` at base `c1bc7e78f3a9debf4f517fb6b5f11dd67be4ad92`. Its successful exact-head checks are stacked-branch integration evidence only; the current protected `main` is `542ca1e9c0a863595b8b6697790005d2471f5413`.
-
-#### Historical exact-head active PR evidence: 2026-08-26
-
-The following newest slices were re-fetched from GitHub for this snapshot. Their exact base/head pairs are recorded so later checks, reviews, and restacks cannot be confused with predecessor evidence:
-
-| PR | State | Exact base head | Exact head |
-|---|---|---|---|
-| #220 | Ready | `b05d5acca82b9d916ada2c8e82f59f92a89817e1` | `e0740a6f3a41067a4460249378e0266815018a74` |
-| #219 | Ready | `b05d5acca82b9d916ada2c8e82f59f92a89817e1` | `3e34a54ae279686a28309d59b8b3b9bfbd283a80` |
-| #218 | Ready | `b05d5acca82b9d916ada2c8e82f59f92a89817e1` | `911ea33d8a5aca7673307bb6fdcad4b450f5c111` |
-| #209 | Ready | `b05d5acca82b9d916ada2c8e82f59f92a89817e1` | `b35d739017aa5d361b605be48045be50b5a35f6f` |
-| #208 | Ready | `b05d5acca82b9d916ada2c8e82f59f92a89817e1` | `e41d3be4c290c4e434aac33d777e511dfb94e03d` |
-| #124 | Ready | `b05d5acca82b9d916ada2c8e82f59f92a89817e1` | `296ad25bb541023dbc869ae07ae1d853820f83a4` |
-
-These rows are delivery evidence only. None has counted independent approval in the current collaborator inventory, and predecessor rows from earlier snapshots are retained below as regression anchors that must never be promoted to current-head evidence.
-
-#### Historical regression-anchor exact-head evidence: superseded 2026-08-24 rows
-
-The following rows were current on 2026-08-24 and are retained only as regression anchors; every listed head has since been superseded or merged and must never be promoted to current-head evidence:
-
-| PR | State | Exact base head | Exact head |
-|---|---|---|---|
-| #222 | Draft | `56fcfa56525e4f2e980e0ee05b6776d621bcddc5` | `1e2ce3d4071a1a75ee891bdcd71c506b3b50d4bc` |
-| #221 | Draft | `8145d40f1b028a8f4dc7e7da47ac89bb9e5bb2c7` | `6f339df1e5b3ddb265f4ddd7b262d4de1e0b5e1f` |
-| #220 | Ready | `0841d2ab3d8b5e60a03c0a8e818cf438e2716829` | `ed4cab16cf88c76ce1c145a22d0a274ef2d57263` |
-| #219 | Ready | `0841d2ab3d8b5e60a03c0a8e818cf438e2716829` | `8145d40f1b028a8f4dc7e7da47ac89bb9e5bb2c7` |
-| #218 | Ready | `0841d2ab3d8b5e60a03c0a8e818cf438e2716829` | `49e98fba6974219b3bb0336c822b12667f1e1c03` |
-| #216 | Ready | `0841d2ab3d8b5e60a03c0a8e818cf438e2716829` | `75130851a0f7ce528a7a36382eb026ac7942a0aa` |
-| #214 | Draft | `40d642d5470a7753b8211907c190367f742f2f12` | `f79999681866ecf0e5fe17d895170f3f6cae7361` |
-| #211 | Draft | `85cc477688246900697f4cfb91c0c8f1f692934a` | `40d642d5470a7753b8211907c190367f742f2f12` |
-| #210 | Draft | `c38b9665774d6b3754e572bed527737b5e179833` | `529d11a3571f6b1834b9baa49ef67eb08f043978` |
-| #209 | Ready | `0841d2ab3d8b5e60a03c0a8e818cf438e2716829` | `c38b9665774d6b3754e572bed527737b5e179833` |
-| #208 | Ready | `0841d2ab3d8b5e60a03c0a8e818cf438e2716829` | `85cc477688246900697f4cfb91c0c8f1f692934a` |
-
-The stack topology shows #209 → #210 → #217 → #222 (WARC/PROV chain, with #217 merged into #210's unprotected branch), #208 → #211 → #214 (BAP chain), and #218 → #221 → #220 (release/enterprise chain) at this snapshot. Every active row above remains PR evidence; none is protected-main behavior.
-
-### Required-check provider failure record
-
-On 2026-08-23 the required Strix security scan failed closed on exact heads of #220 (`ed4cab16…`), #218 (`49e98fba…`), and #208 (`85cc4776…`) because its LLM provider/backend was unavailable (rate limit, token cap, connection, warm-up, or model-behavior failure); no vulnerability report artifact was produced, so the workflow correctly refused to convert an incomplete scan into passing security evidence. Failed jobs were re-dispatched on the unchanged exact heads on 2026-08-24 and again on 2026-08-26. This is a provider-infrastructure failure record, not a weakening of the fail-closed gate or a substitute for a completed authoritative scan.
-
-On 2026-08-26 rerun outcomes were verified per run: completed reruns returned `success` on the heads of #46, #48, #156, #157, #159, #218, and #219; several earlier runs for #37, #43, and #149 were cancelled only because conflict-reconciliation pushes created newer heads with fresh scans; remaining reruns were still in flight at snapshot time. One rerun (#124) produced a real MEDIUM finding (vuln-0001) instead of provider noise; that finding was remediated on the branch head rather than suppressed, preserving the fail-closed contract.
+Active PRs are implementation/review evidence; none of them is protected-main behavior until merged through the current rules. The historical #195/#198 and #149 anchors below remain useful domain examples, while their exact old heads and measurements live only in the evidence dossier.
 
 #### #195/#198 WebDriver BiDi opening path status
 
-Phase 1 is **in progress**, not shipped. #195 and #198 provide bounded WebSocket opening-path evidence on active branches; framed BiDi commands, authenticated browser-process provenance, semantic task execution, and protected-main integration remain open.
+The WebDriver BiDi foundation remains a dependency boundary rather than a shipped-product shortcut. Phase 1 is **in progress**, not shipped. Browser acceptance still requires a real browser session, observed page effects, causal provenance and cleanup evidence on the exact candidate generation.
 
 #### #149 VPN/profile intent status
 
-PR #149 is a ready (non-draft) pull request whose conflict reconciliation and rustfmt correction landed on head `54f96008` on 2026-08-26; it still only describes bounded WireGuard/IKEv2 profile authority and does not create a tunnel, route, DNS state, authenticated gateway, or connectivity proof.
-
-The current queue must be processed in dependency order. A green child branch cannot substitute for current checks and review on its prerequisite, synthetic merge, or eventual protected-main commit. PRs that only duplicate, supersede, or preserve stale branch topology should be closed with explicit replacement evidence rather than retained indefinitely; this loop exercised that policy by closing superseded #153 with replacement evidence.
+The profile/VPN contract remains an intent boundary: it does not create a tunnel, route, DNS state, authenticated gateway, or connectivity proof. Network authority belongs to its canonical owner and must be consumed through an explicit released contract rather than inferred from browser configuration.
 
 ### Review and merge authority
 
-The active `CWL Central required workflows` ruleset (re-fetched for this snapshot) requires one approving review, resolved review threads, no last-push approval requirement, `merge`/`squash` merge methods, and seven configured required workflows (`close-empty-pr`, `opencode-review`, `pr-review-merge-scheduler`, `security-scan`, `strix`, `sast-semgrep`, and `noema-review`). The current collaborator inventory contains only `seonghobae` with administration and push permissions, creating a **reviewer-provisioning gap** for counted non-author approval.
-
-This gap does not authorize self-approval, stale-head merges, administrative bypass, or weaker checks. Because the current GitHub ruleset independently requires a counted approval, the solo-maintainer hold does not satisfy the live merge gate: an eligible non-author collaborator must submit a formal `APPROVED` review on the current head. Until that reviewer-provisioning gap is repaired, protected-main merges stop even when exact-head checks, security gates, complete coverage, rustdoc/Clippy, threads, and AI-review evidence are otherwise complete. Before any merge decision, re-fetch the exact ruleset, collaborators, PR head/base, reviews, unresolved threads, and required checks; do not assume this dated observation remains current.
-
-### Open issues and governance signals
-
-This snapshot contains 11 open issues plus 2 governance signals, for the 13 rows below. Governance signals remain visible because they affect delivery authority but are not counted as product or operational issues.
-
-| Issue or signal | Current gap or signal |
-|---|---|
-| #28 | First real Chromium Agent Task vertical slice; highest immediate Phase 1 buyer-visible gap |
-| #27 | Complete Manifest V3 compatibility and extension-authority isolation matrix |
-| #9 | Bounded HTTP/1.1 semantics over the authenticated TLS stream |
-| #10 | Purpose-bound operational PII disclosure and trusted broker/storage lifecycle |
-| #123 | Fleet incident: disable orphaned TLS, HTTP, and one-shot workflow identities |
-| #187 | Manual-authority review of the coverage-diagnostics workflow delta |
-| #212 | Governance: remove or independently authorize the PR #43 MV3 workflow mutation — **option (b) executed 2026-08-26** with owner-directed authorization recorded on the issue and the mutation restored on the reconciled branch; re-evaluate if the authorization record is contested |
-| #215 | Governance: restore an enforceable protected-main policy that does not create a routine admin bypass |
-| #199 | Schema-bound extraction with durable WARC/PROV replay, retention, deletion, and offline verification |
-| #200 | Stable BAP/MCP runtime API with authenticated, idempotent, cancellable, resumable task lifecycle |
-| #201 | Signed cross-platform Chromium distribution, installer/updater, patch SLA, rollback, SBOM, and provenance |
-| #202 | Enterprise control and experience plane: operator UI, Keyverse-compatible identity, tenancy, approval, audit, SLO, Figma, and Storybook |
-| #203 | Release-grade web-agent benchmark and commercial acceptance gate bound to exact signed artifacts |
-
-Issue #206 (harden-runner custom detection initialization failure) was closed after its remediation landed on protected `main` between snapshots.
-
-The five newly separated product-completion tracks are **durable WARC/PROV replay**, **stable BAP/MCP runtime API**, **signed cross-platform Chromium distribution**, **enterprise control and experience plane**, and the **commercial acceptance gate**. They are separate issues because each has a distinct authority, data, release, and buyer-acceptance boundary.
-
-The hourly product-development loop is operational infrastructure, not proof that a browser product, issue, pull request, or release meets buyer acceptance.
-
-## Buyer-visible and technical gap matrix
-
-| Priority | Buyer-visible outcome | Protected-main status | Completion issue and acceptance evidence |
-|---|---|---|---|
-| P0 | A bounded task observes a real Chromium page, performs one typed action, verifies the post-condition, and emits provenance | **Open / Phase 1** | #28; repeated real Chromium E2E with isolated context, exact session/node authority, typed dispatch, post-condition, crash cleanup, and protected-main checks |
-| P0 | Navigation consumes approved origin, resolution, route, TCP peer, TLS identity, bounded HTTP, redirect, MIME, and download policy | **Partial foundation** | #9 plus #28; real browser-network adapter proves the governed path is consumed end to end |
-| P1 | Existing Chromium extensions remain compatible while Agent authority stays separate | **Partial active-PR evidence** | #27; exact supported-build/platform compatibility matrix, managed allow-list, native-host isolation, repeatability, and release binding |
-| P1 | Authorized work can use necessary PII without ambient exposure | **Policy foundation; runtime open** | #10; opaque broker, exact field/purpose/destination/model policy, atomic use/revocation, retention/deletion, and value-free telemetry |
-| P1 | Every released structured field is traceable to replayable source evidence | **Foundations only** | #199; durable WARC/PROV replay, integrity, retention, deletion, offline verification, extraction precision/recall, and 100% provenance completeness |
-| P1 | External Agents integrate through a stable, authenticated product contract | **Partial active-PR MCP primitives** | #200; BAP 1.0, MCP 2026-07-28 adapter, idempotency, task cancellation/resume, checkpoint/reconciliation, and SDK conformance |
-| P1 | Buyers can install, update, verify, and roll back a supported product | **Not shipped** | #201; signed Windows/macOS/Linux/headless artifacts, Chromium revision manifest, updater security, patch SLA, SBOM, SLSA provenance, and recovery |
-| P1 | Enterprise teams can provision, approve, audit, operate, and recover the service | **Not shipped** | #202; Keyverse-compatible OIDC/SCIM, tenant isolation, policy/approval/evidence UI, SLO/incident controls, data residency, CSAP/SOC 2 evidence mapping, WCAG 2.2, Figma File ID, and Storybook |
-| P0 | A release has reproducible proof of usefulness, safety, evidence completeness, and recovery | **No product-wide release gate** | #203; deterministic, compatibility, adversarial, recovery, and enterprise suites with statistical reporting and an exact-artifact commercial acceptance gate |
-| P0 | Valid changes reach protected `main` without authority improvisation or unbounded stack growth | **Blocked / high integration debt** | Shrink the open-PR queue in dependency order, provision legitimate review authority, require exact-current evidence, and close duplicates/superseded branches |
-
-## Commercial completion definition
-
-OriginWeave is not complete merely because every low-level primitive exists in some open branch. A release candidate is commercially complete only when all of the following are true for the declared support profile:
-
-1. #9, #10, #27, and #28 are integrated on protected `main` as a complete browser/network/action/evidence chain.
-2. #199 provides replayable, retention-governed evidence for every released structured result.
-3. #200 exposes a stable authenticated runtime API and task lifecycle without raw Chromium authority leakage.
-4. #201 produces signed, updateable, rollback-capable release artifacts bound to Chromium, SBOM, and provenance.
-5. #202 supplies tenant-safe enterprise administration, approvals, audit, SLOs, incident recovery, accessible Figma/Storybook-backed UX, and control evidence.
-6. #203 accepts the exact signed artifacts through a reproducible benchmark; missing or inconclusive evidence cannot be promoted to success.
-7. Production function, line, region, and branch coverage and public API documentation remain exactly complete for OriginWeave-owned code.
-8. CHANGELOG, version, supported-platform matrix, security policy, runbooks, licensing, release notes, upgrade/rollback guidance, and procurement evidence match the exact release.
-9. No required check, browser/platform lane, security case, benchmark case, or independent review is skipped, stale, inherited, or represented by status-only evidence.
-10. The open PR queue is reduced to bounded active work rather than being the only place where the product exists.
-
-## Historical next executable queue
-
-1. Drain the merge gate in dependency order: for every ready root PR whose current head is check-green with resolved threads, obtain the current ruleset's counted `APPROVED` review from an eligible non-author collaborator; OpenCode approval or skip evidence does not substitute for that GitHub review. If no eligible approver exists, record the reviewer-provisioning gap and do not merge. Root candidates include #37, #40, #43, #45–#48, #51, #62–#65, #74, #82, #124, #149, #152, #156–#166, #173, #175, #208, #209, #211, #218, #219, #229, #237, #238, and #239 as their current checks land. Treat dependent children separately: only after a predecessor reaches protected `main`, retarget and independently revalidate its immediate child; preserve orders such as #218 → #221 → #220 rather than treating #208–#220 as a flat merge range.
-2. Keep the organization review pipeline healthy: monitor the central Actions backlog recorded above; if OpenCode reviews stop landing on OriginWeave heads while the queue is idle, repair `ContextualWisdomLab/.github` dispatch/concurrency configuration rather than weakening any gate.
-3. Finish the #9/#28 browser-network and Chromium vertical slice, including the #181–#205 WebSocket opening path and framed BiDi command/response stack, then semantic observation, policy, action, post-condition, and recovery boundaries on protected `main`.
-4. Finish #27 and #10 as separate security tracks; neither should be hidden inside the first browser PR.
-5. Implement #199, then #200, so durable evidence and stable task authority precede broad enterprise integrations.
-6. Implement #201 before making release/support claims; exact CI browser evidence must be bound to the actual signed artifact.
-7. Design #202 in Figma, record the Figma File ID in the ADR, implement reusable design tokens and Storybook components, then add identity/tenant/approval/audit/operations integration.
-8. Make #203 the final release gate across the exact signed distribution, not a source branch or model narrative.
-9. Only after the commercial acceptance gate passes, increment the version, finalize CHANGELOG/release notes, publish signed artifacts, and verify upgrade/rollback from the prior supported release.
-
-## Evidence commands
-
-Run `scripts/ci/collect_live_merge_evidence.sh` to collect reusable, exact-head and exact-base merge evidence before interpreting the volatile inventory below.
-
-The executable [collect_live_merge_evidence.sh](../scripts/ci/collect_live_merge_evidence.sh) procedure paginates the complete open-PR and open-issue inventories, excludes pull requests from the issue count, and binds each merge verdict to the exact head, base, checks, reviews, and review threads. It is the canonical procedure; do not copy it into this baseline.
+Ready means review/execution admission, not merge readiness. Re-fetch exact head/base, current reviews, unresolved threads and required checks before any protected merge. Administrative bypass, self-approval, synthesized status, gate weakening, force-push and destructive rebase are not acceptance mechanisms.
 
-The branch-scoped rules response determines the active rules affecting `main`; each PR's exact `HEAD_SHA` then determines which check runs, legacy statuses, workflow runs, reviews, and unresolved threads are current. The saved merge verdict binds counted approvals to the latest review per eligible collaborator, excludes the PR author, and requires `APPROVED` on the exact head. It deliberately does **not** infer GitHub's actual last-push actor from commit author or committer metadata: when `require_last_push_approval` is active, this portable evidence procedure records `github_rule_evaluation_required` and keeps `approval_gate_satisfied` false until GitHub's authoritative rule evaluation is consulted. The saved PR JSON also preserves the exact base reference and branch ancestry input for the dependency graph. Evidence is retained only when both `RECHECKED_HEAD_SHA` and `RECHECKED_BASE_SHA` match the collected values; a moving head or base discards the temporary verdict, and three failed attempts leave no unstable merge verdict.
+## Maintenance rule
 
-For standards and binding architecture, use [`doctoring.md`](doctoring.md), [`doctoring/browser-agent-protocols.md`](doctoring/browser-agent-protocols.md), [`PRD.md`](PRD.md), [`TRD.md`](TRD.md), [`product-roadmap.md`](product-roadmap.md), and linked ADR/UML/ERD/traceability records. Issues #199-#203 contain their own APA 7th standards and research traceability. This baseline intentionally records delivery state and never promotes planned adapters or active pull-request code to implemented behavior.
+Keep this file decision-sized. A new dated observation may replace the bounded observation receipt and matrix evidence, but detailed per-run prose, repeated historical queue dumps, screenshots, long command transcripts and superseded exact-head narratives go to a linked immutable file under `docs/evidence/` or `docs/traceability/`. The regression in `tests/test_product_gap_baseline_navigation_contract.py` fails if the buyer matrix is buried again, a volatile queue snapshot is presented as continuously live, or historical daily cuts are reintroduced here.
diff --git a/docs/traceability/product-gap-baseline-navigation.md b/docs/traceability/product-gap-baseline-navigation.md
new file mode 100644
index 000000000..e60ea2a23
--- /dev/null
+++ b/docs/traceability/product-gap-baseline-navigation.md
@@ -0,0 +1,67 @@
+# Product-gap baseline navigation traceability
+
+- Status: Implemented on active documentation successor; protected-main acceptance pending.
+- Scope: `docs/product-technical-gap-baseline.md` buyer navigation, dated observation semantics, and historical evidence retention only.
+- Predecessor: #238 branch exact `16098a0eecc57bfc545d22ce632d95d1ffccfe67`.
+- Preserved predecessor baseline blob: Git blob `c4a87e6c70f2c48b6e4ef820f18e8ddb0dca5696`, 186,500 bytes.
+- Triggering buyer-navigation finding: #238 comment `5598004000`.
+- Self-staleness finding: #309 review `5154249954` on exact `187310346c5c42b22e53f734c2c9ce7122df0d7f`.
+- Self-staleness RED: `32b01ebda4ce061e63c926434d4320847c93fbcf`.
+- Minimal dated-receipt repair: `84340f2ac35ddbc0571038418acc93e0f9882595`.
+
+## Problem
+
+The baseline had accumulated current observations, repeated same-day checkpoints, old exact heads, measurements, browser screenshots and long implementation runbooks in one 186,500-byte file. The latest observation was accurate, but the commercial gap matrix was no longer a practical decision surface: a buyer or maintainer had to traverse historical evidence before reaching the current gap/owner/acceptance view.
+
+That is a documentation correctness issue, not cosmetic cleanup. When current delivery truth and historical evidence use the same unbounded surface, stale exact-head prose becomes easier to mistake for current acceptance and the next causal action is harder to identify.
+
+The first compact successor exposed a second correctness defect. It labelled a static queue snapshot `Current exact observation` and recorded 129 open PRs before #309 itself existed. Creating #309 changed the queue, and restoring #309 to Draft changed the Ready/Draft split again. A static Markdown file cannot truthfully promise continuously current volatile GitHub counts; self-modifying queue membership makes that claim especially unstable.
+
+## Constraints
+
+- Do not discard or rewrite historical evidence to make the current file shorter.
+- Do not turn active-PR evidence into protected-main or release truth.
+- Preserve exact heads, measurements and dated claims for their recorded generation.
+- Do not present a volatile GitHub snapshot as continuously current after its observation cut.
+- If the documentation PR itself affects the observed queue, include that PR and its lifecycle state in the cut.
+- Do not touch production browser/runtime code, `.github/**`, rulesets, secrets, review state or another bounded context's authority.
+- Preserve existing historical regression coverage rather than deleting assertions that become inconvenient after extraction.
+- Keep the canonical buyer path `docs/product-technical-gap-baseline.md` stable so PRD/TRD/ARCHITECTURE/docs indexes do not need a new authority name.
+
+## Alternatives considered
+
+1. **Keep appending to one file.** Rejected because it preserves evidence but does not repair decision-surface reachability.
+2. **Delete old checkpoints.** Rejected because it destroys traceability and makes previous exact-head claims unauditable.
+3. **Collapse old prose with HTML details while retaining it in the same source file.** Rejected as the final design because rendered navigation improves but source-level history and current authority remain coupled.
+4. **Keep volatile queue counts under a `Current exact` heading.** Rejected because the claim becomes stale as soon as the queue changes and can be invalidated by the documentation PR itself.
+5. **Move the historical dossier to immutable evidence and use an explicitly dated, self-inclusive observation receipt at the canonical path.** Selected. It separates current decision authority from historical receipts, preserves the buyer entry point, and makes the freshness boundary explicit.
+
+## Decision
+
+The predecessor baseline blob is reused byte-for-byte at `docs/evidence/product-technical-gap-baseline-through-2026-09-09.md`. The canonical baseline becomes a bounded dated observation receipt, buyer gap matrix, acceptance order and compact evidence/history index.
+
+Review `5154249954` is repaired test-first. Commit `32b01ebda4ce061e63c926434d4320847c93fbcf` requires the live surface to use an explicit `2026-09-09 12:57 UTC` observation cut, reject the continuously-current heading, include #309's Draft state in the 130-PR / 12-Ready / 118-Draft count, and state that later live GitHub state supersedes the receipt. On predecessor `187310346c5c42b22e53f734c2c9ce7122df0d7f`, those assertions are source-semantic RED. Commit `84340f2ac35ddbc0571038418acc93e0f9882595` makes the smallest documentation repair without changing the archived dossier, production source, workflows, rules, or browser evidence.
+
+Legacy tests whose purpose is to validate historical checkpoint integrity execute against the archived dossier. They are retained as byte-identical non-discovered modules and exposed through small compatibility wrappers that redirect only their `BASELINE` input. The current buyer surface receives its own navigation/freshness regression. This avoids weakening historical assertions while preventing them from forcing historical runbooks back into the live decision surface.
+
+## Risk and mitigation
+
+- **Risk: archive divergence.** Mitigation: the archive reuses the predecessor Git blob rather than reserializing it.
+- **Risk: dated observation mistaken for a live dashboard.** Mitigation: the heading carries the observation time, the text says it is not continuously live, and later GitHub state explicitly supersedes the cut.
+- **Risk: self-reference drift.** Mitigation: the observation cut includes #309 as Draft and the contract pins the complete 130 / 12 / 118 split observed after that lifecycle repair.
+- **Risk: old tests silently stop running.** Mitigation: original `test_*.py` entry points remain discoverable and import/export the preserved legacy test classes after binding their historical baseline input to the archive.
+- **Risk: compactness hides material gaps.** Mitigation: the matrix keeps buyer risk, exact evidence, canonical owner/acceptance and state together, while linked PRs/issues and the evidence dossier retain detail.
+
+## Acceptance
+
+- `docs/product-technical-gap-baseline.md` is at most 220 lines and 24,000 UTF-8 bytes.
+- `## Buyer gap matrix` appears within the first 90 lines.
+- The live baseline contains no `### Previous verified cut:` daily-history section and no `## Current exact observation` heading.
+- The dated observation receipt records `2026-09-09 12:57 UTC`, includes #309 as Draft, records 130 open PRs / 12 Ready / 118 Draft, and says later live GitHub state supersedes the cut.
+- The archived dossier remains larger than 180,000 bytes and contains the known 2026-09-08 and 2026-08-29 historical anchors.
+- Existing historical regression classes remain discoverable through their original test entry points and run against the archived baseline rather than being deleted or relaxed.
+- Protected-main, browser/runtime source, workflows, review rules and external owner contracts remain unchanged by this documentation repair.
+
+## Follow-up
+
+Fresh exact-head hosted checks and rendered GitHub inspection are still required before this documentation successor can be considered accepted. After normal integration into #238, later refreshes should replace the bounded dated receipt/matrix with a new observed cut and add detailed evidence receipts only when that detail no longer belongs on the buyer surface.
diff --git a/tests/_historical_baseline_contract_loader.py b/tests/_historical_baseline_contract_loader.py
new file mode 100644
index 000000000..9d5e42d09
--- /dev/null
+++ b/tests/_historical_baseline_contract_loader.py
@@ -0,0 +1,33 @@
+"""Load preserved historical baseline contracts against extracted evidence receipts."""
+
+from __future__ import annotations
+
+import importlib.util
+from pathlib import Path
+from types import ModuleType
+
+TESTS = Path(__file__).resolve().parent
+ROOT = TESTS.parent
+ARCHIVE = ROOT / "docs/evidence/product-technical-gap-baseline-through-2026-09-09.md"
+ARCHIVE_CHANGELOG = ROOT / "docs/evidence/CHANGELOG-through-2026-09-09.md"
+
+
+def load_historical_contract(filename: str, module_name: str) -> ModuleType:
+    """Load one legacy module against immutable baseline/changelog evidence inputs."""
+    path = TESTS / filename
+    spec = importlib.util.spec_from_file_location(module_name, path)
+    if spec is None or spec.loader is None:
+        raise RuntimeError(f"cannot load historical contract: {filename}")
+    module = importlib.util.module_from_spec(spec)
+    spec.loader.exec_module(module)
+    module.BASELINE = ARCHIVE
+    if hasattr(module, "CHANGELOG"):
+        module.CHANGELOG = ARCHIVE_CHANGELOG
+    return module
+
+
+def export_public(module: ModuleType, namespace: dict[str, object]) -> None:
+    """Expose the preserved test classes/helpers through the original discovery module."""
+    for name in dir(module):
+        if not name.startswith("_"):
+            namespace[name] = getattr(module, name)
diff --git a/tests/legacy_documentation_active_pr_evidence_contract.py b/tests/legacy_documentation_active_pr_evidence_contract.py
new file mode 100644
index 000000000..a0d2ff3d0
--- /dev/null
+++ b/tests/legacy_documentation_active_pr_evidence_contract.py
@@ -0,0 +1,333 @@
+"""Regression contracts for volatile active-PR evidence in canonical documentation."""
+
+from pathlib import Path
+import re
+import unittest
+
+
+ROOT = Path(__file__).resolve().parents[1]
+DOCS = ROOT / "docs"
+FITNESS = DOCS / "DOCUMENTATION_FITNESS.md"
+MATURITY = DOCS / "evidence" / "2026-08-10-active-pr-maturity.md"
+BASELINE = DOCS / "product-technical-gap-baseline.md"
+CHANGELOG = ROOT / "CHANGELOG.md"
+
+
+def active_pr_row(text: str, pr_number: int) -> str:
+    """Return exactly one maturity row for an active pull request."""
+    prefix = f"| #{pr_number} |"
+    rows = [line for line in text.splitlines() if line.startswith(prefix)]
+    if len(rows) != 1:
+        raise AssertionError(
+            f"expected exactly one active maturity row for PR #{pr_number}, got {len(rows)}"
+        )
+    return rows[0]
+
+
+def bounded_section(text: str, start: str, end: str) -> str:
+    """Return one explicitly bounded documentation section, failing closed on drift."""
+    if start not in text:
+        raise AssertionError(f"missing section start marker: {start}")
+    remainder = text.split(start, 1)[1]
+    if end not in remainder:
+        raise AssertionError(f"missing section end marker after {start}: {end}")
+    return remainder.split(end, 1)[0]
+
+
+class ActivePullRequestDocumentationContractTests(unittest.TestCase):
+    """Keep volatile implementation evidence separate from protected-main truth."""
+
+    @classmethod
+    def setUpClass(cls) -> None:
+        cls.fitness = FITNESS.read_text(encoding="utf-8")
+        cls.maturity = MATURITY.read_text(encoding="utf-8")
+        cls.baseline = BASELINE.read_text(encoding="utf-8")
+        cls.changelog = CHANGELOG.read_text(encoding="utf-8")
+
+    def test_latest_live_pr_snapshot_is_recorded_in_the_product_baseline(self) -> None:
+        """The current section must preserve exact heads for the newest active slices."""
+        current = bounded_section(
+            self.baseline,
+            "#### Current exact-head active PR evidence",
+            "#### Historical 2026-08-26 maintenance-loop record",
+        )
+        for marker in (
+            "| #73 | Draft | `600d3975c02b68da1974a4c73069b966b39dce7b` | `ce1b138509ab4f52cb0f80290f104358473c6ed3` |",
+            "| #72 | Draft | `f86ce504138e79d6e95141a441f60b40920e1fa6` | `600d3975c02b68da1974a4c73069b966b39dce7b` |",
+            "| #46 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `373113119446d99f578febd39efc19366e7736b1` |",
+            "| #70 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `441a8ce1d09c329c5c1168f4906d9a38fd0abc01` |",
+            "| #82 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `f5776f5f233ac0a7c05e3f4a2846436c23438043` |",
+            "| #152 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `81407a0e5189a413d1be0963fea90a0c2f254ce1` |",
+            "| #210 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `7946dce9a3dd074047d93fca299d48c7aef40e47` |",
+            "| #237 | Draft | `542ca1e9c0a863595b8b6697790005d2471f5413` | `2459af602e72fbfe1ce816919473a1075ec0c41f` |",
+            "| #239 | Draft | `e45cd6cdcdee73b5c16dc942e6c98cb7e745fae0` | `e840ca299d29a15223c8b9bb1397002c4f41b4a3` |",
+            "| #229 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `0145ccba5901e301b41d4be674ca1ed23483ad37` |",
+            "| #220 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `b11db2be68f9b6d71aa4c4290b97a8b22097b353` |",
+            "| #211 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `52a918577958a5701e1146c7eb8b62fe8f8ccd44` |",
+            "| #195 | Draft | `6922dd98779e8f8aad132a3b1f563d7ba6e6d070` | `48eb2d23009c1c804520dd5efcd0d4d072aacef1` |",
+            "| #242 | Draft | `48eb2d23009c1c804520dd5efcd0d4d072aacef1` | `55fef0c3fae1724eddada53e52c4a0311f509aa3` |",
+            "| #93 | Draft | `802ec806cdd4560eab48c484f435766ecabda353` | `0664f0452cb329cd692cce7f61f9001652abfda2` |",
+            "| #95 | Draft | `0664f0452cb329cd692cce7f61f9001652abfda2` | `97aa0f2e340ee6fd920d0418f97af276b190554f` |",
+            "| #96 | Draft | `97aa0f2e340ee6fd920d0418f97af276b190554f` | `b7ba8dd1433410cee43084a73e31816da841b2a2` |",
+            "| #101 | Draft | `b7ba8dd1433410cee43084a73e31816da841b2a2` | `bc810b121bb0303f55afa8777a23cc0f9748c1db` |",
+            "| #102 | Draft | `bc810b121bb0303f55afa8777a23cc0f9748c1db` | `a123c55d4839dae1db7e6671f7d4d158c7cfd9db` |",
+            "| #103 | Draft | `a123c55d4839dae1db7e6671f7d4d158c7cfd9db` | `8b3416169346fa04b53c915b813d55ccf47d1876` |",
+            "| #124 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `fdb88698ca20626a6643bc2ad7944fb968835700` |",
+            "| #37 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `5e3dfcbd7a4daea297782cb99635990368589232` |",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, current)
+        self.assertNotIn("| #238 |", current)
+        self.assertIn("PR #238 itself", current)
+
+    def test_baseline_refresh_changelog_matches_the_live_snapshot(self) -> None:
+        """The current changelog refresh item must carry its own live queue evidence."""
+        added = self.changelog.split("### Added", 1)[1].split("### Changed", 1)[0]
+        changed = self.changelog.split("### Changed", 1)[1].split("### Security", 1)[0]
+        refresh_prefix = "- Revalidated the product-gap queue at"
+        refresh_lines = [line for line in added.splitlines() if line.startswith(refresh_prefix)]
+        self.assertEqual(1, len(refresh_lines))
+        refresh_line = refresh_lines[0]
+        latest_cut = bounded_section(
+            self.baseline,
+            "### Latest verified cut: 2026-09-09",
+            "### Previous verified cut: 2026-09-08",
+        )
+        current = latest_cut
+        queue_counts = re.findall(
+            r"\*\*(\d+) open pull requests: (\d+) Ready/non-draft and (\d+) Draft; "
+            r"(\d+) open non-PR issues\*\*",
+            " ".join(current.split()),
+        )
+        self.assertEqual(1, len(queue_counts))
+        total, ready, draft, issues = queue_counts[0]
+        inventory_lines = [
+            line for line in self.changelog.splitlines()
+            if line.startswith("- Current delivery inventory:")
+        ]
+        self.assertEqual(1, len(inventory_lines))
+        self.assertIn(f"{total} open pull requests ({ready} ready, {draft} draft)", inventory_lines[0])
+        self.assertIn(f"{issues} open non-PR issues", inventory_lines[0])
+        self.assertIn("Observed 2026-09-09", inventory_lines[0])
+        self.assertIn(
+            "Revalidated the active ruleset inventory at 7 required workflows",
+            changed,
+        )
+        self.assertIn("`codeql-pr`", changed)
+        self.assertIn("Made the open non-PR issue count reproducible", changed)
+        self.assertNotIn("- Corrected the 2026-08-28 product-gap snapshot", added)
+        self.assertNotIn("- Revalidated the product-gap queue at", changed)
+        self.assertNotIn("115 open pull requests (31 ready, 84 draft)", refresh_line)
+        self.assertNotIn("126 open pull requests (54 ready, 72 draft)", refresh_line)
+        self.assertNotIn("128 open pull requests (54 ready, 74 draft)", refresh_line)
+        self.assertNotIn("153 open pull requests (39 ready, 114 draft)", refresh_line)
+
+    def test_latest_inventory_drift_cannot_be_hidden_by_historical_counts(self) -> None:
+        """Changing only the newest count must invalidate an unchanged changelog."""
+        latest = bounded_section(
+            self.baseline,
+            "### Latest verified cut: 2026-09-09",
+            "### Previous verified cut: 2026-09-08",
+        )
+        self.assertIn("131 open pull requests", latest)
+        mutated_latest = latest.replace(
+            "131 open pull requests",
+            "132 open pull requests",
+            1,
+        )
+        self.assertNotEqual(latest, mutated_latest)
+        probe = ActivePullRequestDocumentationContractTests(
+            "test_baseline_refresh_changelog_matches_the_live_snapshot"
+        )
+        probe.baseline = self.baseline.replace(latest, mutated_latest, 1)
+        probe.changelog = self.changelog
+        with self.assertRaises(AssertionError):
+            probe.test_baseline_refresh_changelog_matches_the_live_snapshot()
+
+    def test_current_warc_provider_failures_are_bound_to_current_head(self) -> None:
+        """WARC evidence must describe the current parent head after stack merge."""
+        for marker in (
+            "PR #210 current exact head is `7946dce9a3dd074047d93fca299d48c7aef40e47`",
+            "`Rust contracts` job `98942518975` and `Production coverage` job `98942518680` succeeded",
+            "`noema-review` job `98942513421` and `strix` job `98942803402` remain in progress",
+            "Its predecessor head `5f59947f5e4b0d3bc0aa5b2d4c6722d3b7c43047`, prior stack merge head `66f360ccac5cec60c72222cc79d58e39f6f00088`, earlier exact head `bea65643109449d63d367a35b8d9bf327ee7cb2c`, and their OpenCode/Strix provider failures remain historical evidence only",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, self.baseline)
+
+    def test_current_opencode_dispatch_failures_are_bound_to_current_heads(self) -> None:
+        """OpenCode dispatch authorization failures must remain explicit blockers."""
+        for marker in (
+            "A direct central `opencode-review` dispatch run `33192478312` / job `98921183278` was rejected",
+            "The immediately preceding PR #238 head `d0b0d1ed92f891f14646fc673b8e1c0d912586fd` remains historical",
+            "automatic OpenCode run `33193822920` / job `98926243116` failed closed without a current-head verdict",
+            "central dispatch run `33194506918` / job `98928580387` also failed closed at OpenCode",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, self.baseline)
+
+    def test_dependency_stacks_are_explicit_and_non_shipped(self) -> None:
+        """Current browser, network, sensitive and compatibility stacks stay active-only."""
+        for pr_number in (52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66):
+            with self.subTest(pr_number=pr_number):
+                row = active_pr_row(self.maturity, pr_number)
+                self.assertIn("**IMPLEMENTED_ON_ACTIVE_PR**", row)
+                self.assertNotIn("IMPLEMENTED_ON_PROTECTED_MAIN", row)
+
+        for stack in (
+            "#47 → #50 → #54",
+            "#45 → #46 → #53 → #55",
+            "#40→#52→#57→#58",
+            "#43→#56→#59→#60→#61",
+            "#51→#66",
+        ):
+            with self.subTest(stack=stack):
+                self.assertIn(stack, self.fitness)
+
+        self.assertIn("#49", self.fitness)
+
+    def test_semantic_relationship_evidence_stays_bounded_and_authority_scoped(self) -> None:
+        """PR #52 cannot turn relationship metadata into browser or execution authority."""
+        row = active_pr_row(self.maturity, 52)
+        for marker in ("128", "relationship", "session/context/origin/document"):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, row)
+
+        for marker in (
+            "same browser session, browsing context, canonical origin and document epoch",
+            "Self-parent/self-child relationships and duplicate child handles fail closed",
+            "relationship graph remains descriptive evidence",
+            "not a browser observation adapter",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, self.fitness)
+
+    def test_typed_semantic_query_evidence_stays_descriptive_and_bounded(self) -> None:
+        """PR #57 cannot turn semantic matching into selector or execution authority."""
+        row = active_pr_row(self.maturity, 57)
+        for marker in (
+            "SemanticNodeQuery",
+            "role",
+            "accessible-name",
+            "typed-action",
+            "no CSS/XPath/raw DOM selector language",
+            "browser I/O or action authority",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, row)
+
+        self.assertIn("Draft stacked on exact #52 head", row)
+        self.assertIn("CI run `31429995885`", row)
+        self.assertIn("CodeRabbit exact-head status succeed", row)
+        self.assertIn("remains Draft because #52/#40 are active prerequisites", row)
+
+    def test_sensitive_audience_evidence_does_not_claim_authentication(self) -> None:
+        """An internal audience field is not authenticated workload/service identity."""
+        row = active_pr_row(self.maturity, 55)
+        self.assertIn("authenticated workload/service identity", row)
+        self.assertIn(
+            "audience string accepted by the value/policy primitive is **not authentication**",
+            self.fitness,
+        )
+        self.assertIn("new deployment topology or physical ERD entity", self.fitness)
+
+    def test_mv3_mutation_and_isolation_are_compatibility_not_agent_authority(self) -> None:
+        """Real MV3 evidence must remain separate from OriginWeave capability grants."""
+        bookmark_row = active_pr_row(self.maturity, 56)
+        for marker in ("create", "get", "remove", "compatibility evidence only"):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, bookmark_row)
+
+        for pr_number in (59, 60, 61):
+            with self.subTest(pr_number=pr_number):
+                row = active_pr_row(self.maturity, pr_number)
+                self.assertIn("**IMPLEMENTED_ON_ACTIVE_PR**", row)
+                self.assertNotIn("IMPLEMENTED_ON_PROTECTED_MAIN", row)
+
+        self.assertIn("Manifest V3 compatibility", self.fitness)
+        self.assertIn(
+            "Chromium permission or browser compatibility success is not an OriginWeave Agent capability",
+            self.fitness,
+        )
+        self.assertIn(
+            "#43/#49/#56/#59/#60/#61 are active compatibility evidence only",
+            self.fitness,
+        )
+        self.assertIn("Update migration is intentionally distinct from restart persistence", self.fitness)
+        self.assertIn("isolated-world behavior is intentionally distinct from injection alone", self.fitness)
+
+    def test_extension_proposal_grant_does_not_become_agent_policy_authority(self) -> None:
+        """PR #62 must remain a policy-isolation regression, not a new action grant."""
+        row = active_pr_row(self.maturity, 62)
+        for marker in (
+            "ProposeTypedAction",
+            "out-of-grant target origin",
+            "missing core `Navigate` capability",
+            "untrusted instruction source",
+            "adds no production API or real Chromium adapter",
+            "does not convert extension proposal authority into Agent action/origin authority",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, row)
+        self.assertIn("CI run `31436844685`", row)
+        self.assertIn("Security Scan run `31436844615`", row)
+        self.assertIn("SAST Semgrep run `31436844646`", row)
+        self.assertNotIn("IMPLEMENTED_ON_PROTECTED_MAIN", row)
+
+    def test_latest_agent_task_and_secret_composition_evidence_remains_partial(self) -> None:
+        """Newest active slices must not be promoted into a complete browser or broker runtime."""
+        secret_approval = active_pr_row(self.maturity, 63)
+        for marker in (
+            "ProposeTypedAction",
+            "RequireApproval(RiskClass::R3)",
+            "no secret broker",
+        ):
+            with self.subTest(pr_number=63, marker=marker):
+                self.assertIn(marker, secret_approval)
+
+        action_outcome = active_pr_row(self.maturity, 64)
+        for marker in (
+            "PostConditionPredatesDispatch",
+            "monotonic",
+            "not a browser dispatcher",
+        ):
+            with self.subTest(pr_number=64, marker=marker):
+                self.assertIn(marker, action_outcome)
+
+        controlled_fixture = active_pr_row(self.maturity, 65)
+        for marker in (
+            "controlled",
+            "prompt-injection",
+            "not a browser adapter",
+        ):
+            with self.subTest(pr_number=65, marker=marker):
+                self.assertIn(marker, controlled_fixture)
+
+        process_set = active_pr_row(self.maturity, 66)
+        for marker in (
+            "process-set RSS",
+            "duplicate",
+            "does not discover Chromium PIDs",
+        ):
+            with self.subTest(pr_number=66, marker=marker):
+                self.assertIn(marker, process_set)
+
+        for marker in (
+            "#62/#63",
+            "#64",
+            "#65",
+            "#51→#66",
+            "real Chromium",
+        ):
+            with self.subTest(fitness_marker=marker):
+                self.assertIn(marker, self.fitness)
+
+    def test_erd_stays_conceptual_without_persistence_owner(self) -> None:
+        """Active in-memory/value primitives must not manufacture a physical data model."""
+        self.assertIn("Conceptual ERD/domain model", self.fitness)
+        self.assertIn("add no OriginWeave-owned durable store", self.fitness)
+        self.assertIn("false architecture", self.fitness)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/legacy_gap_snapshot_inventory_consistency.py b/tests/legacy_gap_snapshot_inventory_consistency.py
new file mode 100644
index 000000000..d4c9c0599
--- /dev/null
+++ b/tests/legacy_gap_snapshot_inventory_consistency.py
@@ -0,0 +1,461 @@
+"""Regression contracts for current and historical product-gap inventory evidence."""
+
+from __future__ import annotations
+
+from pathlib import Path
+import re
+import unittest
+
+
+ROOT = Path(__file__).resolve().parents[1]
+BASELINE = ROOT / "docs" / "product-technical-gap-baseline.md"
+CHANGELOG = ROOT / "CHANGELOG.md"
+
+
+class GapSnapshotInventoryConsistencyTests(unittest.TestCase):
+    """Keep volatile live truth separate from immutable dated snapshots."""
+
+    @classmethod
+    def setUpClass(cls) -> None:
+        cls.baseline = BASELINE.read_text(encoding="utf-8")
+        cls.changelog = CHANGELOG.read_text(encoding="utf-8")
+
+    def test_current_live_state_is_distinct_from_dated_snapshot(self) -> None:
+        """The live section must bind fresh evidence without promoting queued work."""
+        current = self.baseline.split("## Current live delivery state", 1)[1].split(
+            "## Observed snapshot: 2026-08-29", 1
+        )[0]
+        for marker in (
+            "125 open pull requests",
+            "12 Ready/non-draft",
+            "113 Draft",
+            "13 open non-PR issues",
+            "87c4daa1830bac5a5228b6036752ad5633232085",
+            "18156473",
+            "7 central required workflows",
+            "codeql-pr",
+            "Live GitHub PR/base/head/check APIs are authoritative over PR bodies",
+            "Ready roots are #37 `1e2f41072854edcdbaf0f9ecf14697a3bfd62195`, #50 `ad87cfea59db711cb29ef90559790ba77e22029f`, #166 `e84a1a2cc82b1c666218efd441da97849f47b8c2`, #219 `65e4315d80137badc0b55e1b9617015beb1db568`, #220 `e545b94e1de499b96b867694f80ac04ad247becd`, #229 `024f63690cf05cfe6f0d4a430f0e18ea8fd2c4d6`, #240 `24930a3a9ee79c0b712ee3df6589b0592eb6e18f`, #272 `b1cae8ad1cbd8eb6992037c830aea30b9aa436b3`, #274 `802d0bdff7536d9ac253305d3e0237b4e4a1789e`, #285 `f455c2cd64b3dd3f027c91d396103792a205ddd0`, and #287 `af83c40dd2990a03064a92ca75430a9cc400f098`",
+            "PR #290 is Draft at exact head `ebeefcd534db4324498fdb18046ebc6255ddcdf2`",
+            "PR #238's moving exact head is intentionally omitted",
+            "PR #166 exact head `e84a1a2cc82b1c666218efd441da97849f47b8c2`",
+            "PR #220 exact head `e545b94e1de499b96b867694f80ac04ad247becd`",
+            "retain formal `CHANGES_REQUESTED` decisions",
+            "thread-resolution state could not be refreshed",
+            "115 queued workflow runs",
+            "rerunning unchanged heads would only add duplicate queue load",
+            "PR #249 is Draft at exact head `84b9407978ae0f6c115f01170b6069c601b21104`",
+            "989 functions, 9978 lines, 12717 regions, 1098 branches",
+            "PR #250 now adopts #249 at `ec433b844a121f8554c062f92267991af9cacb6f`",
+            "1023 functions, 10518 lines, 13525 regions, 1186 branches",
+            "PR #251 now adopts #250 at `f02af6d0dd01708d495cc08dec785675f3d58898`",
+            "1037 functions, 10647 lines, 13681 regions, 1192 branches",
+            "PR #252 now adopts #251 at `6569bf40b6595ac74c2f0a997d202137f07ba1db`",
+            "1044 functions, 10699 lines, 13751 regions, 1192 branches",
+            "PR #253 now adopts #252 at `afb623e4449b7cbf926fdcef7225ceaca822cfcf`",
+            "1054 functions, 10747 lines, 13787 regions, 1194 branches",
+            "Review `5120013340` remains unresolved",
+            "PR #254 now adopts #253 at `b11b6c9bccd8335b58a4fb599f8ad29ac419637f`",
+            "1061 functions, 10802 lines, 13843 regions, 1200 branches",
+            "Review `5120077272` remains actionable",
+            "Current #148 `bded4fc9d32e5e047cea99d182aa05ae2cd03bf6`",
+            "all 247 Python contracts pass on Linux with zero skips",
+            "Current #150 `d3c29359fd10e540ca9b3b2723bde6f94866cdb6`",
+            "all 229 Python contracts pass on macOS and Linux",
+            "#166 Strix run `33929688857` / job `101237371800` failed",
+            "console classifier also matches ordinary denied-policy prose",
+            "#219 Strix run `33925442296` / job `101226800608`",
+            "six-hour GitHub job execution limit",
+            "#240 Strix run `33925596890` / job `101227893828`",
+            "Current #147 `3dff28d9bf2dd27b72507e39979d51b8bf140fb4`",
+            "all six focused and all 228 native tests now pass",
+            "all 228 native tests execute but two forced-close protocol-fault subcases fail",
+            "Temporary integration PR #291 is now closed",
+            "PR #142 at `015025f2539e4fb1dbd7d259ec22dad50f944396`",
+            "187 Python contracts passed",
+            "PR #143 at `44fd9a450f864feff5cf2ba2883425a71ba10b9b`",
+            "192 Python contracts",
+            "415 functions, 3555 lines, 4444 regions, 476 branches",
+            "process-observation errors leave termination unproven",
+            "PR #144 at `09f2e087d0c20fe81386c18099e739a9e611a8ad`",
+            "194 Python contracts passed",
+            "separate root and process-set deadline budgets",
+            "Current PR #144 parent adoption is `3c0c5c363c2da0b4e8a621226e2f7766c735b743`",
+            "all 196 Python contracts",
+            "PR #145 now adopts #144 at `bb5e8f834c37f9ce35f84db8ed1146da3659d6aa`",
+            "all 202 Python contracts",
+            "PR #146 review repair is `812c0020bd2ecdb3eda39d999ed3327647550bdf`",
+            "10 assertion failures and 7 uncaught protocol-error cases",
+            "all 224 Python contracts",
+            "concurrent parent-adoption `11944410450684809ee1a71a35c77abafc5358db`",
+            "Noema job `101245089068` failed with HTTP 502 after 2739.2 seconds",
+            "Strix scan job `101243872506` executed successfully",
+            "#250 `0eab23d5e388c5c8b984c0021a58316680c9ba8b`",
+            "#251 `86e8ad76838f2a64aa7e0cd56ba1f931c8d0c3dc`",
+            "#252 `2015259529ada99af836989079cc85a15779a2d8`",
+            "#253 `0d72082e595c0e1fcc03d609ba337896ed14e2fc`",
+            "three unauthenticated caller booleans mint `OperationallyComplete`",
+            "raw teardown claims must remain explicitly unverified",
+            "#254 `cbaf50dcc97753cc73135497ea8225e8b18de190`",
+            "#255 `a13de5f9321e72c1867974eb7a43230f031e58df`",
+            "Follow-up review `5120203295` at #255 head `92392b2da33bcd5446c6d2eb1b5504c39e60e3d6`",
+            "received-message provenance must be checked before correlation is consumed",
+            "Received-connection verification at #255 head `e7bfec4488b7cb4776df7b546cacb46c8c9eb13e`",
+            "12 focused tests passed",
+            "lines=11046/11049 and regions=14067/14071",
+            "Follow-up quality verification at #255 head `63cbca0a98cf9496af981819d98029e656fc4342`",
+            "functions=1082/1083, lines=11046/11051, regions=14071/14074",
+            "unused private accessor",
+            "Prior #255 source-owner repair is `ebac126d1632c94775c2454423575275eec45def`",
+            "PR #255 now adopts #254 at `3e7057443d7c9532ff526acb5eefe8cd4778c767`",
+            "PR #256 now adopts #255 at `ced4a851ca66c08d895a098725c7f0ad3ecf0c38`",
+            "PR #257 now adopts #256 at `f4a8f2cbf515bea348f500b59615a9581c1b96a2`",
+            "1082 functions, 11032 lines, 14086 regions, 1202 branches",
+            "1088 functions, 11077 lines, 14146 regions, 1210 branches",
+            "1093 functions, 11131 lines, 14202 regions, 1214 branches",
+            "1100 functions, 11185 lines, 14272 regions, 1214 branches",
+            "1140 functions, 11841 lines, 15129 regions, 1332 branches",
+            "1154 functions, 11969 lines, 15314 regions, 1334 branches",
+            "Central Strix #1563 advanced to `13fbb48e0b3eeca4ce7d9678add934f9bd87ad3f`",
+            "independently inspected four-file source diff",
+            "2890 passed, 1 skipped, and 21 subtests are owner-reported",
+            "521 functions, 4420 lines, 5349 regions, 654 branches",
+            "#37 Strix run `33946242342`",
+            "Strix run timed out after 900s",
+            "#148 hosted compatibility run `33962062608`",
+            "browser-crash recovery fails 0/3 trials",
+            "#261 now includes actual #260",
+            "1173 functions, 12104 lines, 15511 regions, 1334 branches",
+            "1221 functions, 12781 lines, 16404 regions, 1418 branches",
+            "no subscription bytes but two outstanding commands",
+            "GraphQL rejected both the inventory query and a minimal probe",
+            "REST reviews, check runs, and combined statuses were read for all 125 open PRs",
+            "warning: --branch option is unstable",
+            "numerical coverage enforcement is distinct from warning-free instrumentation",
+            "PR #139 at `b7ea5bfe336456fb263dd479a60b1cd0193d8a47`",
+            "Opening-exchange fixture repair at #242",
+            "PR #243 adopts that corrected parent by ordinary merge",
+            "PR #246 adopts #243 at `585791f3641fbe757c3bd9fd36d5316adcc78d63`",
+            "PR #248 is Draft at `b386f17c4826adabebda084bff2fba35aee94dd0`",
+            "native unittest discovery collected zero checks",
+            "975 functions, 9848 lines, 12563 regions, 1092 branches",
+            "962 functions, 9701 lines, 12427 regions, 1084 branches",
+            "PR #141 at `fbdf64f5818ce0c53b475196d5bcfa2ac9900846`",
+            "183 Python contracts passed",
+            "two informational threads were resolved without source changes",
+            "Noema attempt 2 is queue-admission evidence, not provider recovery or a review verdict",
+            "central CodeQL dispatches remain queued and were not duplicated",
+            "#256 `9f2e6f29be46371762e3031a97c1cac04720694f`",
+            "#257 `ea2b5b78868917219c46f1304558b92490a7f6fe`",
+            "Issue #279",
+            "Issue #28 remains the P0 governed-browser integration target",
+            "PR #260 is Draft at exact head `2c5049aff97a90958e8262b1d403bdcbd64a1e8b`",
+            "PR #258 is Draft at exact head `5f830324f6d5a47ac213a57528ef95649bdfd0df`",
+            "PR #259 is Draft at exact head `66731982ed51ad62a04fcfbc759b0a33721a0254`",
+            "PR #261 is Draft at exact head `934eb7d37568b439c442ffe1d1f6a9c8f8ed58a0`",
+            "Repair PR #277 is Draft at exact head `117f6414e8a6db46eb2b32f4ebae85cf2a208371`",
+            "PR #70 is Draft at exact head `77eb0f2ee71783e06171784b7173c0b4cd530e61`",
+            "DDD/MCP repair #272 is Ready at exact head `b1cae8ad1cbd8eb6992037c830aea30b9aa436b3`",
+            "PR #229 is Ready at exact head `024f63690cf05cfe6f0d4a430f0e18ea8fd2c4d6`",
+            "PR #281 remains Draft at exact head `adaca6427d68f550b39293a69b7c733430d1c385`",
+            "PR #282 is Draft at exact head `b64e0708584beff3fb54acf226cb3e667773e473`",
+            "PR #283 is Draft at exact head `c904300a6a1bda83af24f84d586f1c5f6a6491aa`",
+            "stacked on predecessor #282 head `b54a5856d8201911f05d69622f0d5594a371adf0` rather than current #282 exact `b64e0708584beff3fb54acf226cb3e667773e473`",
+            "must adopt the corrected current parent before its compare can become current evidence",
+            "PR #288 is Draft at exact head `0f434bc29d468127412366b6864b733b40b83c4d`",
+            "all 173 Python contracts pass on that unchanged head",
+            "PR #269 is Draft at exact head `7854394266d3f292e779193c01413a34f6798d7c`",
+            "PR #270 is Draft at exact head `191a14535219ea8033777fa4c970efb281b62418`",
+            "PR #271 is Draft at exact head `802ec806cdd4560eab48c484f435766ecabda353`",
+            "PR #285 is Ready at exact head `f455c2cd64b3dd3f027c91d396103792a205ddd0`",
+            "CI `33930234387`",
+            "Rust contracts `101207153048` and Production coverage `101207153244` succeeded",
+            "native CI success does not replace the seven required central workflows",
+            "Security Scan `33924016851`",
+            "SAST Semgrep `33924016903`",
+            "CodeQL PR `33924016883`",
+            "did not materialize fresh central required-workflow runs",
+            "PR #287 is Ready at exact head `af83c40dd2990a03064a92ca75430a9cc400f098`",
+            "#287 CodeQL handoff RCA",
+            "33954721186",
+            "33955024697",
+            "33955164029",
+            "PR #273 is Draft at exact head `e5c8fcb66bf644dfa750bb1b40ba3d600cb7805a`",
+            "GitHub Releases is empty",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, current)
+        self.assertNotRegex(
+            current,
+            re.compile(r"PR #238 exact head `[0-9a-f]{40}`"),
+            "The self-referential baseline cannot pin the SHA produced by its own commit; read #238 live metadata instead.",
+        )
+        self.assertRegex(
+            current,
+            re.compile(r"Observed at \(UTC\): `\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}Z`"),
+        )
+        for non_passing in (
+            "queued reviewer evidence is non-passing",
+            "discard queued, skipped, cancelled, absent, predecessor, synthetic, status-only, and model-only evidence",
+        ):
+            with self.subTest(non_passing=non_passing):
+                self.assertIn(non_passing.casefold(), current.casefold())
+        self.assertIn("binds Git rename/copy similarity to blob identity", current)
+        self.assertNotIn("remaining rename/copy similarity-binding RED", self.changelog)
+        for stale in (
+            "Protected `main` is `c789b802fc98a8d7fd8c09d9327f36828054d2a1` through #280",
+            "143 open pull requests",
+            "142 open pull requests",
+            "24 Ready/non-draft",
+            "138 Draft",
+            "121 Draft",
+            "12 open non-PR issues",
+            "10 central required workflows",
+            "6 central required workflows",
+            "PR #260 is Draft at exact head `a3741389fdc491c7ecccc20f77609c55bc56d20f`",
+            "PR #261 remains stale-parent Draft at exact head `127e02503e48938e29a9a07410574c7e72fc661a`",
+            "DDD/MCP repair #272 is Draft at exact head `80272f18422c9946077ad9bd674f603db8f020da`",
+            "PR #229 is Draft at exact head `7ae426e760e8351ee792ce9df4266d7e7483d0d4`",
+            "PR #282 is Draft at exact head `b2a120f892973e76c0ea0f06e7105bdf7a268009`",
+            "PR #283 is Draft at exact head `a9603170e848a7c029531fe75727f02992ade2de`",
+            "PR #282 is Draft at exact head `e3a40a4f78a3fbfc751ab9efad321b8207fb43e5`",
+            "PR #283 is Draft at exact head `85d31596c8b7251135f773b1d54d0f656fa10bbf`",
+            "PR #260 is Draft at exact head `56600a6fd982cfafd784f4b7bb659d918113ca90`",
+            "PR #281 remains Draft at exact head `3ed5d7e8cf77547c96feff2cfb24c46d74a73ebb`",
+        ):
+            with self.subTest(stale=stale):
+                self.assertNotIn(stale, current)
+
+    def test_latest_cut_does_not_promote_prior_observations(self) -> None:
+        """Fresh queue evidence must precede the explicitly historical cut."""
+        latest = self.baseline.split("## Current live delivery state", 1)[1].split(
+            "### Prior observation cut: 2026-09-05", 1
+        )[0]
+        for marker in (
+            "Latest verified cut: 2026-09-06",
+            "63997bcf555e2c5c8e91ba287734ffba3837a1b7",
+            "fd589cd693946ef1ce2c9270c2dfb6a1087bdfb9",
+            "0135984f1bc1f68d89d7777f49c4999474105a12",
+            "973e34bc24ae9bdd96a50764f2be6c8603eff66c",
+            "3f22de94b63da83eaa8b5b1270912b21a3ecd006",
+            "GraphQL thread resolution was refreshed for all 125 open PRs",
+            "#147 retains the sole unresolved thread",
+            "34013251657",
+            "33990522248",
+            "34009256997",
+            "#279 owns current-workflow reconstruction",
+            "#212 owns sandbox-helper adoption",
+            "earlier local coverage counts do not validate these newer heads",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, latest)
+
+    def test_current_cut_records_parent_adoption_and_sent_command_repair(self) -> None:
+        """New source evidence must not inherit earlier native or browser acceptance."""
+        latest = self.baseline.split("## Current live delivery state", 1)[1].split(
+            "### Prior observation cut: 2026-09-05", 1
+        )[0]
+        for marker in (
+            "6ab7fc9166f397f49442243492ae88d6ccad56cf",
+            "43d3b5a3a2b5ce4f51a93d1152a0ee82620f4f3e",
+            "73f11de2232060ac7680e188db88ef7609123296",
+            "253 Python contracts",
+            "1273/13294/16963/1430",
+            "34020055803",
+            "actual-resend freshness",
+            "current caller-supplied registry",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, latest)
+
+    def test_current_cut_separates_browser_failure_and_unpublished_repair(self) -> None:
+        """Hosted browser failure and private source proof must stay distinct."""
+        latest = self.baseline.split("## Current live delivery state", 1)[1].split(
+            "### Prior observation cut: 2026-09-05", 1
+        )[0]
+        for marker in (
+            "9985228944",
+            "05b6d1cd6092a9e2e18a5003a5a1faeb73b8a8d9b4fd7533f7f24bc938833436",
+            "5557994915",
+            "bd29f405a6c927b2f7d1c437dccbfb8bcec424f1",
+            "1273/13311/16973/1432",
+            "unpublished checkpoint",
+            "5557964460",
+            "writer acknowledgement",
+            "8ebcc6131a5dd6bf0b4720c2f1ff8d40d1cc39f8",
+            "lines 13327/13330 and regions 17004/17008",
+            "functions 1274/1274 and branches 1434/1434",
+            "f352e3f04bdc3ed1912be7c452c5c426d9caaec7dd1a613b7d3ab9c6486ab69e",
+            "formatting check fails in five test/support files",
+            "Neither private nor remote results establish combined-source acceptance",
+            "2a9fdc5418b9af10353cdad0f6f6470655bf457d",
+            "52cff96954c3fec7b8cda5409e4560e970bfcbdf",
+            "1273/13317/16984/1432",
+            "f7e71e6fd67723688535053c389b4ff718c4563b403924af462848d3d51b541f",
+            "34024499232",
+            "5558261278",
+            "both complete lineages are now ancestors",
+            "too_many_active_automations",
+            "10f138f8787d596e8b556fe50c9e4e52bc1295b7",
+            "b3ffeac9",
+            "1278/13371/17056/1434",
+            "9f5b4942d6a040a83f09593a2b81406c44bc9d2b402fa995a95294ce8e9cbb7e",
+            "34026519860",
+            "34026519878",
+            "owner destruction",
+            "registry-to-transport association",
+            "805051527cf95e14ba126c9dd3159db86d190224",
+            "1279/13409/17087/1436",
+            "64a62b1e03ee3ed3d62654a3227495e82b7ff93357038ae146d8584c581ac060",
+            "34029687813",
+            "34029687816",
+            "terminal-success predecessor evidence",
+            "actual Edge visual inspection",
+            "GraphQL access recovered",
+            "7fb93e77b800f27187a5c02333298cc31a025bd6",
+            "1279/13421/17106/1438",
+            "277c5bc0a17f16966a8eb388d54e4e172be36356d7dead5bbead9ed53ad8cff6",
+            "34031281825",
+            "34031281812",
+            "new-head visual inspection remains incomplete",
+            "Mac is locked",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, latest)
+        self.assertNotIn(
+            "Native CI `34018588292` and MV3 `34018588298` remain queued",
+            latest,
+        )
+
+    def test_pointer_integration_cut_separates_parent_and_hosted_acceptance(self) -> None:
+        """Published pointer proof must retain the newer parent and visual evidence gaps."""
+        latest = self.baseline.split("#### Current pointer integration and independently owned parent repair", 1)[1].split(
+            "#### Scheduling and central-owner follow-up", 1
+        )[0]
+        for marker in (
+            "35555d0f8491d4ee95c2e61d1a2aaa2c02a0635c",
+            "3f9cdc1a",
+            "1293/13567/17273/1440",
+            "b1374f65538fb6557b4e5d877aa067e183202ed1f56c75be047b3ff2dfdc92a0",
+            "34032661775",
+            "34032661781",
+            "b4702cd503fa3f721e0d1f44b355563753dac0a2",
+            "5559076006",
+            "not latest-parent acceptance",
+            "four silently deleted parent postcondition tests",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, latest)
+        self.assertIn("Browser access recovered", latest)
+        self.assertIn("published body and Checks page", latest)
+        self.assertIn("new documentation head still requires its own visual inspection", latest)
+        self.assertIn("5559140277", latest)
+
+    def test_session_adoption_cut_records_executed_and_visual_evidence(self) -> None:
+        """Current source evidence must not inherit earlier pending-adoption claims."""
+        latest = self.baseline.split("### Latest verified cut: 2026-09-06", 1)[1].split(
+            "#### Prior observation: 12:28 UTC", 1
+        )[0]
+        for marker in (
+            "7147893c96ca95c9b5b275d8011c5bfe99aab065",
+            "6f331a5b220349a1aaa1b1841d5e8ec027b9ad49",
+            "98621adf", "1296/13601/17303/1442",
+            "480d6d3ca492565e8a04bfce4b0c135dffbff276532d129c410674838e8fec2f",
+            "34036335342", "34036335505", "queued",
+            "actual Edge visual inspection", "not product-browser acceptance",
+            "5559497861", "not protected-main delivery",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, latest)
+
+    def test_current_baseline_inventory_matches_the_verified_snapshot(self) -> None:
+        """The dated 2026-08-29 snapshot must keep its exact historical inventory."""
+        current = self.baseline.split("### Open pull requests", 1)[1].split(
+            "#### 2026-08-29 maintenance-loop record", 1
+        )[0]
+        for marker in (
+            "108 open pull requests",
+            "24 non-draft",
+            "84 draft",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, current)
+
+        for stale in (
+            "128 open pull requests",
+            "126 open pull requests",
+            "54 non-draft",
+            "72 draft",
+            "115 open pull requests",
+            "31 non-draft",
+            "116 open pull requests",
+            "32 non-draft",
+            "111 open pull requests",
+            "27 non-draft",
+            "153 open pull requests",
+            "114 draft PRs",
+        ):
+            with self.subTest(stale=stale):
+                self.assertNotIn(stale, current)
+
+    def test_unreleased_changelog_uses_one_current_inventory(self) -> None:
+        """The current Unreleased entry must match the verified volatile inventory."""
+        unreleased = self.changelog.split("## [Unreleased]", 1)[1]
+        preamble, remainder = unreleased.split("### Added", 1)
+        added = remainder.split("### Changed", 1)[0]
+
+        expected = "125 open pull requests (12 ready, 113 draft)"
+        self.assertIn(expected, preamble)
+        self.assertIn("13 open non-PR issues", preamble)
+        self.assertIn(expected, added)
+        self.assertIn("13 open non-PR issues", added)
+        self.assertNotIn("143 open pull requests", preamble)
+        self.assertNotIn("108 open pull requests (24 ready, 84 draft)", preamble)
+
+    def test_current_snapshot_records_recent_stack_merges_and_revalidation(self) -> None:
+        """A stacked merge must update the dated queue and parent exact-head evidence."""
+        current = self.baseline.split("### Open pull requests", 1)[1].split(
+            "#### 2026-08-29 maintenance-loop record", 1
+        )[0]
+        record = self.baseline.split(
+            "#### 2026-08-29 maintenance-loop record", 1
+        )[1].split("#### Current exact-head active PR evidence", 1)[0]
+        self.assertIn("108 open pull requests", current)
+        self.assertIn("24 non-draft", current)
+        self.assertIn("#217 was squash-merged", record)
+        self.assertIn(
+            "#53 was exact head `4ecc81e59ae7bc3a640e65e2442bf30c079bd94c`",
+            record,
+        )
+        self.assertIn(
+            "#217 was exact head `6b8a3fdeae52ad94b90086bbc9b42863b90c9614`",
+            record,
+        )
+        self.assertIn("66f360ccac5cec60c72222cc79d58e39f6f00088", record)
+        self.assertIn("#67 was squash-merged", record)
+        self.assertIn("5021d142583cb5a8e393248048bb824762a98056", record)
+        self.assertIn("PR #64 consequently advanced", record)
+        self.assertIn(
+            "| #210 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | "
+            "`7946dce9a3dd074047d93fca299d48c7aef40e47` |",
+            self.baseline,
+        )
+        self.assertNotIn("| #217 |", current)
+
+    def test_current_documentation_head_records_security_and_review_state(self) -> None:
+        """The dated self-referential record must preserve its historical gate truth."""
+        record = self.baseline.split(
+            "#### 2026-08-29 maintenance-loop record", 1
+        )[1].split("#### Current exact-head active PR evidence", 1)[0]
+        for marker in (
+            "The immediately preceding PR #238 head `d0b0d1ed92f891f14646fc673b8e1c0d912586fd` remains historical",
+            "automatic OpenCode run `33193822920` / job `98926243116` failed closed",
+            "current Strix run `33193822929` / job `98925769697` succeeded",
+            "central dispatch run `33194506918` / job `98928580387` also failed closed at OpenCode",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, record)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/legacy_live_gap_evidence_integrity_contract.py b/tests/legacy_live_gap_evidence_integrity_contract.py
new file mode 100644
index 000000000..998be0669
--- /dev/null
+++ b/tests/legacy_live_gap_evidence_integrity_contract.py
@@ -0,0 +1,171 @@
+"""Fail-closed contracts for the current product-gap evidence procedure."""
+
+from pathlib import Path
+import subprocess
+import unittest
+
+
+ROOT = Path(__file__).resolve().parents[1]
+BASELINE = ROOT / "docs" / "product-technical-gap-baseline.md"
+CHANGELOG = ROOT / "CHANGELOG.md"
+AGENTS = ROOT / "AGENTS.md"
+EVIDENCE_SCRIPT = ROOT / "scripts" / "ci" / "collect_live_merge_evidence.sh"
+
+
+def bounded(text: str, start: str, end: str) -> str:
+    """Return one current section without accepting a historical substitute."""
+    if start not in text:
+        raise AssertionError(f"missing start marker: {start}")
+    remainder = text.split(start, 1)[1]
+    if end not in remainder:
+        raise AssertionError(f"missing end marker after {start}: {end}")
+    return remainder.split(end, 1)[0]
+
+
+class LiveGapEvidenceIntegrityContractTests(unittest.TestCase):
+    """Keep merge evidence bound to the exact current PR state."""
+
+    @classmethod
+    def setUpClass(cls) -> None:
+        cls.baseline = BASELINE.read_text(encoding="utf-8")
+        cls.changelog = CHANGELOG.read_text(encoding="utf-8")
+        cls.agents = AGENTS.read_text(encoding="utf-8")
+        cls.evidence = EVIDENCE_SCRIPT.read_text(encoding="utf-8")
+        cls.latest = bounded(
+            cls.baseline,
+            "### Latest verified cut: 2026-09-09",
+            "### Previous verified cut: 2026-09-08",
+        )
+
+    def test_latest_inventory_and_changelog_use_the_september_9_cut(self) -> None:
+        marker = (
+            "131 open pull requests: 14 Ready/non-draft and "
+            "117 Draft; 14 open non-PR issues"
+        )
+        self.assertIn(marker, " ".join(self.latest.split()))
+        inventory = [
+            line
+            for line in self.changelog.splitlines()
+            if line.startswith("- Current delivery inventory:")
+        ]
+        self.assertEqual(1, len(inventory))
+        self.assertIn("131 open pull requests (14 ready, 117 draft)", inventory[0])
+        self.assertIn("14 open non-PR issues", inventory[0])
+        self.assertIn("Observed 2026-09-09", inventory[0])
+
+    def test_presentation_snapshot_uses_full_exact_sha(self) -> None:
+        full_sha = "6855e2578ae94279cc9ab4a14527b016e8c049ee"
+        self.assertIn(full_sha, self.latest)
+        self.assertNotIn("to exact head\n`6855e257`", self.latest)
+
+    def test_active_presentation_harness_is_not_described_as_unimplemented(self) -> None:
+        normalized = " ".join(self.latest.split())
+        self.assertIn("active controlled evidence harness is implemented", normalized)
+        self.assertIn("failed 0/3 at session creation before navigation", normalized)
+        self.assertIn("product Browser Session adapter", normalized)
+
+    def test_presentation_successor_records_its_repaired_contract_boundary(self) -> None:
+        """Keep the active child repair distinct from browser-runtime acceptance."""
+        normalized = " ".join(self.latest.split())
+        self.assertIn("Draft successor #298", normalized)
+        self.assertIn("d01f45c2c8ac7b0fc4dbc3d3ada60238732cdf8c", normalized)
+        self.assertIn("#305", normalized)
+        self.assertIn("does not establish live Chromium transport", normalized)
+
+    def test_documented_current_evidence_collector_is_executable(self) -> None:
+        current_evidence = self.baseline.split("## Evidence commands", 1)[1]
+        self.assertIn("scripts/ci/collect_live_merge_evidence.sh", current_evidence)
+        self.assertIn("scripts/ci/collect_live_merge_evidence.sh", self.agents)
+        self.assertNotEqual(0, EVIDENCE_SCRIPT.stat().st_mode & 0o111)
+
+    def test_evidence_collector_is_valid_bash(self) -> None:
+        result = subprocess.run(
+            ["bash", "-n", str(EVIDENCE_SCRIPT)],
+            cwd=ROOT,
+            check=False,
+            capture_output=True,
+            text=True,
+        )
+        self.assertEqual(0, result.returncode, result.stderr)
+
+    def test_current_change_requests_block_the_approval_verdict(self) -> None:
+        self.assertIn("as $current_change_requests", self.evidence)
+        self.assertIn("blocking_change_requests: $current_change_requests", self.evidence)
+        self.assertIn("($current_change_requests | length) == 0", self.evidence)
+
+    def test_every_unresolved_thread_remains_blocking_when_outdated(self) -> None:
+        self.assertIn("select(.isResolved == false)", self.evidence)
+        self.assertNotIn(
+            "select(.isResolved == false and .isOutdated == false)",
+            self.evidence,
+        )
+
+    def test_workflow_evidence_is_bound_to_pr_head_and_base(self) -> None:
+        for marker in (
+            ".number == ($pr[0].number)",
+            ".head.sha == $head",
+            ".base.sha == $base",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, self.evidence)
+        self.assertIn("workflow_runs_without_exact_pr_base_provenance", self.evidence)
+
+    def test_rules_are_evaluated_for_each_pull_requests_actual_base(self) -> None:
+        for marker in (
+            "BASE_REF=$(jq -r '.base.ref' \"$PR_JSON\")",
+            'BASE_REF_ENCODED=$(jq -rn --arg value "$BASE_REF" \'$value | @uri\')',
+            '"repos/$REPOSITORY/rules/branches/$BASE_REF_ENCODED?per_page=100"',
+            '--slurpfile rules "$EVIDENCE_DIR/pr-${PR}-branch-rules.json"',
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, self.evidence)
+        self.assertNotIn(
+            '--slurpfile rules "$EVIDENCE_DIR/main-branch-rules.json"',
+            self.evidence,
+        )
+
+    def test_verdict_materializes_only_after_inventory_and_final_state_stabilize(self) -> None:
+        for marker in (
+            "INVENTORY_DRAFT=$(jq -r --argjson pr \"$PR\"",
+            "PR_STATE=$(jq -r '.state' \"$PR_JSON\")",
+            "PR_DRAFT=$(jq -r '.draft' \"$PR_JSON\")",
+            "RECHECKED_STATE=$(jq -r '.state' \"$RECHECKED_PR_JSON\")",
+            "RECHECKED_DRAFT=$(jq -r '.draft' \"$RECHECKED_PR_JSON\")",
+            '"$PR_STATE" == "open"',
+            '"$RECHECKED_STATE" == "$PR_STATE"',
+            '"$RECHECKED_DRAFT" == "$PR_DRAFT"',
+            '"$PR_DRAFT" == "$INVENTORY_DRAFT"',
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, self.evidence)
+
+    def test_whole_generation_closes_against_fresh_pr_and_issue_inventories(self) -> None:
+        for marker in (
+            'open-pr-pages-rechecked.json',
+            'open-prs-rechecked.json',
+            'open-issue-pages-rechecked.json',
+            'open-issues-rechecked.json',
+            'INITIAL_PR_INVENTORY_PROJECTION=',
+            'FINAL_PR_INVENTORY_PROJECTION=',
+            'INITIAL_ISSUE_INVENTORY_PROJECTION=',
+            'FINAL_ISSUE_INVENTORY_PROJECTION=',
+            'Live inventory changed during evidence collection.',
+            'rm -f "$EVIDENCE_DIR"/pr-*-merge-verdict.json',
+            'evidence-generation.json',
+            'complete: true',
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, self.evidence)
+
+    def test_explicit_evidence_directory_must_start_empty(self) -> None:
+        for marker in (
+            'if [[ $# -gt 0 ]]; then',
+            'find "$EVIDENCE_DIR" -mindepth 1 -print -quit',
+            'Evidence directory must be empty:',
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, self.evidence)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/legacy_product_completion_gap_contract.py b/tests/legacy_product_completion_gap_contract.py
new file mode 100644
index 000000000..970c9c189
--- /dev/null
+++ b/tests/legacy_product_completion_gap_contract.py
@@ -0,0 +1,790 @@
+"""Regression contract for the dated commercial-completion gap baseline."""
+
+from __future__ import annotations
+
+import pathlib
+import re
+import unittest
+from unittest.mock import patch
+
+from test_documentation_active_pr_evidence_contract import active_pr_row, bounded_section
+
+ROOT = pathlib.Path(__file__).resolve().parents[1]
+BASELINE = ROOT / "docs/product-technical-gap-baseline.md"
+
+
+class ProductCompletionGapContractTests(unittest.TestCase):
+    """Keep the exact repository snapshot and completion tracks reviewable."""
+
+    def test_ack_checkpoint_separates_local_repair_from_review_acceptance(self) -> None:
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "#### Published intent acknowledgment verification",
+            "#### Published repeated-control repair",
+        )
+        for marker in (
+            "46db0045904f0289738df843d0a2f179c26673d3", "34186280263",
+            "8eda96915dbbe4cc617f834267c7464689c2844d",
+            "1406/14894/19018/1552", "pre-consumption", "resolved",
+            "terminal SUCCESS", "CLEAN", "not shipped", "148 repository",
+            "10d5e1ff78d46bbad004d4e0971d0ffceac757fa", "superseded",
+            "6b6c90ed3919ea84b69527ab688a087eeb45224d",
+            "1105/11574/14787/1218", "34188785932", "server-sent",
+            "RED", "1010", "1011", "Actual Edge", "terminal SUCCESS",
+            "6b5241c164f5283f8dd51b1846ef0e4dacec0b29", "predecessor",
+            "0c077445", "594/4974/5980/754", "#293", "three command intents",
+            "3 September 2026", "viewport/DPR reset intent",
+            "current-head visual acceptance is unproven",
+            "Viewport", "DevicePixelRatio", "TimeZone", "ReducedMotion",
+        ):
+            self.assertIn(marker, current)
+
+    def test_repeated_control_checkpoint_keeps_operational_gaps_open(self) -> None:
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "#### Published repeated-control repair",
+            "#### Published deadline and Close-code repairs",
+        )
+        for marker in (
+            "3a8e6f4f89db2f53c144adb3351c153d89adca58", "34184829970",
+            "1104/11553/14758/1216", "20 closure", "64", "65th",
+            "not hosted acceptance", "process exit", "profile cleanup",
+            "not shipped", "fresh", "exhausted", "reused",
+        ):
+            self.assertIn(marker, current)
+
+    def test_close_code_checkpoint_does_not_claim_hosted_acceptance(self) -> None:
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "#### Published deadline and Close-code repairs",
+            "#### Transport-closure verification and test-integrity repair",
+        )
+        for marker in (
+            "8716b9d441960a112446c1f89ab417ad9abe28d2", "34183766437",
+            "1101/11492/14662/1218", "18 closure", "142 repository",
+            "not hosted acceptance", "repeated Ping/Pong", "1016–2999",
+            "process exit", "profile cleanup", "not shipped",
+        ):
+            self.assertIn(marker, current)
+
+    def test_closure_checkpoint_separates_verified_and_pending_heads(self) -> None:
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "### Previous verified cut: 2026-09-08",
+            "## Observed snapshot: 2026-08-29",
+        )
+        for marker in (
+            "d126242c7198c447d0fab7983d529441340fd1c9",
+            "07ef43ec71b6dbd8540629bf1df5a63b81541ee4",
+            "363a78e36e7690e9ed5bf49829567e00e2ec5d59",
+            "34179452950", "34180304951", "1090/11218/14328/1210",
+            "126 open pull requests", "114 Draft", "14 open non-PR issues",
+            "not protected-main acceptance", "operation-wide deadline",
+            "repeated Ping/Pong", "no package release",
+        ):
+            self.assertIn(marker, current)
+
+    def test_end_reply_checkpoint_binds_published_evidence(self) -> None:
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "#### Published session-end reply binding: 05:35 UTC",
+            "#### Published status-response repair: 04:45 UTC",
+        )
+        for owner, head, parent, coverage in (
+            (251, "924ad97551750d4a901ded38b89488cc5438e54f", "#250 `bbdc6ace`", "1057/10840/13873/1194"),
+            (252, "363a78e36e7690e9ed5bf49829567e00e2ec5d59", "#251 `924ad975`", "1064/10898/13947/1194"),
+        ):
+            row = active_pr_row(current, owner)
+            for marker in (head, parent, coverage, "Published; Draft"):
+                self.assertIn(marker, row)
+        for marker in (
+            "34085877650", "34087239755",
+            "9f2249637f31916acf9874baec724ddedfe621e5b14fdb6f4bd850620e89f307",
+            "not hosted acceptance", "retained receipts", "#255", "#292",
+        ):
+            self.assertIn(marker, current)
+
+    def test_historical_evidence_cannot_replace_end_reply_checkpoint(self) -> None:
+        text = BASELINE.read_text(encoding="utf-8")
+        current = bounded_section(
+            text, "#### Published session-end reply binding: 05:35 UTC",
+            "#### Published status-response repair: 04:45 UTC",
+        )
+        for owner in (251, 252):
+            row = active_pr_row(current, owner)
+            stale = text.replace(row, row.replace("Published; Draft", "Local only"), 1)
+            with patch.object(pathlib.Path, "read_text", return_value=stale + row):
+                with self.assertRaises(AssertionError):
+                    self.test_end_reply_checkpoint_binds_published_evidence()
+
+    def test_status_repair_checkpoint_binds_published_evidence(self) -> None:
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "#### Published status-response repair: 04:45 UTC",
+            "#### Published semantic-action adoption: 04:11 UTC",
+        )
+        for marker in (
+            "bbdc6ace7a5932adf24836700f806850e6b230bc",
+            "Published; Draft",
+            "1043/10711/13717/1188",
+            "34084134654",
+            "e1fccefc6b56eabe653ae41377fbcec0aa89b4be1ada92dba73d3ef87e841029",
+            "not hosted acceptance",
+        ):
+            self.assertIn(marker, current)
+
+    def test_historical_evidence_cannot_replace_status_repair(self) -> None:
+        text = BASELINE.read_text(encoding="utf-8")
+        current = bounded_section(
+            text, "#### Published status-response repair: 04:45 UTC",
+            "#### Published semantic-action adoption: 04:11 UTC",
+        )
+        stale = text.replace(current, current.replace("Published; Draft", "Local only"), 1)
+        with patch.object(pathlib.Path, "read_text", return_value=stale + current):
+            with self.assertRaises(AssertionError):
+                self.test_status_repair_checkpoint_binds_published_evidence()
+
+    def test_current_snapshot_checks_do_not_route_by_phrase_substrings(self) -> None:
+        """Current-snapshot assertions must not depend on count-word substrings."""
+        source = pathlib.Path(__file__).read_text(encoding="utf-8")
+        brittle_condition = '"pull requests" in phrase or ' + '"draft" in phrase'
+        self.assertNotIn(brittle_condition, source)
+
+    def test_baseline_records_current_inventory_and_completion_issues(self) -> None:
+        """The dated baseline must not retain superseded queue counts or omit buyer tracks."""
+        text = BASELINE.read_text(encoding="utf-8")
+        end_marker = "#### 2026-08-29 maintenance-loop record"
+        self.assertIn(end_marker, text)
+        current = text.split("## Observed snapshot: ", 1)[1].split(end_marker, 1)[0]
+
+        for phrase in (
+            "108 open pull requests",
+            "24 non-draft",
+            "84 draft",
+            "2026-08-28 116-PR snapshot",
+        ):
+            with self.subTest(phrase=phrase):
+                self.assertIn(phrase, current)
+
+        for phrase in (
+            "#198",
+            "#199",
+            "#200",
+            "#201",
+            "#202",
+            "#203",
+            "Shrink the open-PR queue",
+            "durable WARC/PROV replay",
+            "stable BAP/MCP runtime API",
+            "signed cross-platform Chromium distribution",
+            "enterprise control and experience plane",
+            "commercial acceptance gate",
+        ):
+            with self.subTest(phrase=phrase):
+                self.assertIn(phrase, text)
+
+        self.assertNotIn("Shrink the 109-PR queue", text)
+        self.assertNotIn(
+            "current-head hosted CI, security, Noema, scheduler, and OpenCode workflows were still regenerating at the recheck",
+            text,
+        )
+
+        for stale_phrase in (
+            "100 open pull requests",
+            "22 non-draft",
+            "78 draft",
+            "148 open pull requests",
+            "79 draft PRs",
+            "40 non-draft",
+            "110 draft",
+            "150 open pull requests",
+            "prior 150-PR snapshot",
+            "128 open pull requests",
+            "74 draft",
+            "115 open pull requests",
+            "31 non-draft",
+        ):
+            with self.subTest(stale_phrase=stale_phrase):
+                self.assertNotIn(stale_phrase, current)
+
+    def test_active_github_approval_rule_is_not_documented_as_bypassable(self) -> None:
+        """An active counted-approval rule must stop merge without an eligible approver."""
+        text = BASELINE.read_text(encoding="utf-8")
+
+        self.assertIn("eligible non-author", text)
+        self.assertIn("reviewer-provisioning gap", text)
+        self.assertNotIn("owner-directed administrative merge", text)
+
+    def test_current_snapshot_records_the_pr_284_admin_bypass_incident(self) -> None:
+        """A bypassed main update must not be presented as a policy-compliant merge."""
+        text = BASELINE.read_text(encoding="utf-8")
+        current = text.split("## Current live delivery state", 1)[1].split(
+            "## Observed snapshot: ", 1
+        )[0]
+
+        for marker in (
+            "87c4daa1830bac5a5228b6036752ad5633232085",
+            "#284",
+            "61bcf88c960c6c437ccd29b3fbb73cd4325f9e5a",
+            "rule-suite `3948421709`",
+            "`result: bypass`",
+            "#215",
+            "fresh native Rust and production-coverage checks now succeed",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, current)
+
+        self.assertNotIn("through #280", current)
+
+    def test_current_snapshot_records_repaired_webdriver_bidi_lineage(self) -> None:
+        """Only the latest cut may establish current foundation and transport heads."""
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "#### Connection-bound text responses: 14:54 UTC",
+            "#### Session repair and child adoption: 13:35 UTC",
+        )
+        self.assertIn("#195 `63997bcf555e2c5c8e91ba287734ffba3837a1b7`", current)
+        self.assertIn("#267 `3346d8ecc72932b98ec495d9cc52d6e5727c3064`", current)
+        self.assertIn("#268 `e567af9e678fd4791776df795e89ed666975e6c2`", current)
+        for marker in (
+            "4632f2df", "d6889c80", "e1188c86",
+            "34040202356", "34040306372", "queued",
+            "pointer and status", "not product-browser acceptance",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, current)
+
+    def test_latest_executable_queue_uses_current_ready_roots(self) -> None:
+        text = BASELINE.read_text(encoding="utf-8")
+        current = bounded_section(
+            text, "#### Published session-end reply binding: 05:35 UTC",
+            "#### Published status-response repair: 04:45 UTC",
+        )
+        roots = [line for line in current.splitlines() if line.startswith("Ready roots:")]
+        self.assertEqual(len(roots), 1)
+        self.assertEqual(
+            set(re.findall(r"#(\d+)", roots[0])),
+            {"37", "50", "166", "219", "220", "229", "238", "240", "272", "274", "285", "287"},
+        )
+        self.assertIn("## Historical next executable queue", text)
+        self.assertNotIn("## Next executable queue", text)
+
+    def test_pointer_checkpoint_records_published_receipt_repair(self) -> None:
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "##### Pointer receipt follow-up: 15:28 UTC",
+            "#### Session repair and child adoption: 13:35 UTC",
+        )
+        for marker in (
+            "#257 `9451fd8a23dec95b31749376bc78c2eaca977fe8`",
+            "#258 `5417ce32ed957aa166807f1023647caccc2920cb`",
+            "8193fcd5", "d9396f05", "588fe731", "0234b587",
+            "2 → 2 → 1", "34041977863", "34042223733", "queued",
+            "not product-browser acceptance", "outbound session authority",
+            "#249", "#250", "descendant adoption",
+        ):
+            self.assertIn(marker, current)
+
+    def test_historical_text_cannot_supply_missing_pointer_receipt_head(self) -> None:
+        text = BASELINE.read_text(encoding="utf-8")
+        marker = "#258 `5417ce32ed957aa166807f1023647caccc2920cb`"
+        mutated = text.replace(marker, "response head removed", 1)
+        self.assertNotEqual(mutated, text)
+        mutated += "\nHistorical evidence: " + marker
+        with patch.object(pathlib.Path, "read_text", return_value=mutated):
+            with self.assertRaises(AssertionError):
+                self.test_pointer_checkpoint_records_published_receipt_repair()
+
+    def test_descendant_checkpoint_records_exact_receipt_adoptions(self) -> None:
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "##### Receipt descendant adoption: 16:17 UTC",
+            "#### Session repair and child adoption: 13:35 UTC",
+        )
+        for marker in (
+            "180c168ecbdcd5eb7a4ad14ab4a53e8670646bf7",
+            "3807aabeb22f9622610c3c8d504d1c686d25d896",
+            "ba100fbc39e1ac4f10ee4faade38418551bb8298",
+            "121578a43adda12221a9ca9ab8ada0a4fd03efef",
+            "34043222194", "34043664230", "34044193429", "34044857929",
+            "queued", "144 Python", "subscription response provenance",
+            "protected-main asset", "not product-browser acceptance",
+        ):
+            self.assertIn(marker, current)
+
+    def test_historical_text_cannot_supply_missing_descendant_head(self) -> None:
+        text = BASELINE.read_text(encoding="utf-8")
+        marker = "121578a43adda12221a9ca9ab8ada0a4fd03efef"
+        mutated = text.replace(marker, "subscription head removed", 1)
+        self.assertNotEqual(mutated, text)
+        mutated += "\nHistorical evidence: " + marker
+        with patch.object(pathlib.Path, "read_text", return_value=mutated):
+            with self.assertRaises(AssertionError):
+                self.test_descendant_checkpoint_records_exact_receipt_adoptions()
+
+    def test_subscription_checkpoint_records_verified_repair_and_visual_gap(self) -> None:
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "##### Subscription response repair: 16:37 UTC",
+            "#### Session repair and child adoption: 13:35 UTC",
+        )
+        for marker in (
+            "46ae62aa31e35c702cd61c16322d05c7a9c35da1", "122ca139", "8b1508c8",
+            "0/2", "14 focused", "144 Python", "1222/12824/16453/1418",
+            "34045953423", "queued", "Mac is locked", "visual inspection remains pending",
+            "outbound session binding", "protected-main asset", "not product-browser acceptance",
+        ):
+            self.assertIn(marker, current)
+
+    def test_observation_safeguards_bind_each_published_owner(self) -> None:
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "#### Published observation safeguards: 03:00 UTC",
+            "#### Published response and observation: 01:45 UTC",
+        )
+        for owner, head, parent, coverage in (
+            (270, "8eda96915dbbe4cc617f834267c7464689c2844d", "3df2a631", "1346/14113/17928/1464"),
+            (271, "b0410ae92bd20eaf31d09b7d49390e13cb045999", "8eda9691", "1393/14770/18899/1546"),
+        ):
+            row = active_pr_row(current, owner)
+            for marker in ("Published; Draft", parent, coverage, f"/commit/{head}"):
+                self.assertIn(marker, row)
+        for marker in (
+            "29cd0d66", "14efb678", "148 Python", "34076117534", "34077987302",
+            "queued", "actual visual inspection", "not product-browser acceptance",
+            "original connection", "status receipts", "#195/#279",
+        ):
+            self.assertIn(marker, current)
+
+    def test_action_adoption_checkpoint_binds_published_owners(self) -> None:
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "#### Published semantic-action adoption: 04:11 UTC",
+            "#### Published observation safeguards: 03:00 UTC",
+        )
+        for owner, head, parent, coverage in (
+            (93, "82056d13aa94c106060b84ee76be56fcb7787fc8", "b0410ae9", "1403/14852/18976/1552"),
+            (95, "6b29d890245ed2f612c2998198f4e8c8a06da312", "82056d13", "1408/14897/19022/1552"),
+            (96, "cbabf55c6a25b979fa0d9e3c1677338665975ab7", "6b29d890", "1410/14908/19030/1552"),
+        ):
+            row = active_pr_row(current, owner)
+            for marker in ("Published; Draft", parent, coverage, f"/commit/{head}"):
+                self.assertIn(marker, row)
+        for marker in (
+            "34079739018", "34080772063", "34081979602", "queued",
+            "actual visual inspection", "not product-browser acceptance",
+            "original connection", "status receipts", "publish = false",
+        ):
+            self.assertIn(marker, current)
+
+    def test_historical_rows_cannot_replace_action_adoption(self) -> None:
+        text = BASELINE.read_text(encoding="utf-8")
+        current = bounded_section(text, "#### Published semantic-action adoption: 04:11 UTC", "#### Published observation safeguards: 03:00 UTC")
+        for owner in (93, 95, 96):
+            row = active_pr_row(current, owner)
+            for replacement in ("", row.replace("Published; Draft", "Local only; Draft")):
+                mutated = text.replace(row, replacement, 1) + "\nHistorical evidence:\n" + row
+                with patch.object(pathlib.Path, "read_text", return_value=mutated):
+                    with self.assertRaises(AssertionError):
+                        self.test_action_adoption_checkpoint_binds_published_owners()
+
+    def test_historical_rows_cannot_replace_observation_safeguards(self) -> None:
+        text = BASELINE.read_text(encoding="utf-8")
+        current = bounded_section(text, "#### Published observation safeguards: 03:00 UTC", "#### Published response and observation: 01:45 UTC")
+        for owner in (270, 271):
+            row = active_pr_row(current, owner)
+            for replacement in ("", row.replace("Published; Draft", "Local only; Draft")):
+                mutated = text.replace(row, replacement, 1) + "\nHistorical evidence:\n" + row
+                with patch.object(pathlib.Path, "read_text", return_value=mutated):
+                    with self.assertRaises(AssertionError):
+                        self.test_observation_safeguards_bind_each_published_owner()
+
+    def test_response_observation_checkpoint_binds_each_published_owner(self) -> None:
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "#### Published response and observation: 01:45 UTC",
+            "#### Published input descendants: 00:40 UTC",
+        )
+        for owner, head, parent, coverage in (
+            (268, "ff27220cb5eb4d11ca1dc5614a4181e1a397a3f1", "ebd507ae", "1325/13907/17682/1456"),
+            (269, "3df2a631bacd7109b3982fdd7ac599d0bd92a589", "ff27220c", "1338/14025/17843/1460"),
+        ):
+            row = active_pr_row(current, owner)
+            for marker in ("Published; Draft", parent, coverage):
+                self.assertIn(marker, row)
+            self.assertIn(f"[`{head[:8]}`](https://github.com/ContextualWisdomLab/OriginWeave/commit/{head})", row)
+        for marker in ("658fb676", "49d18f5f", "147 Python", "34072645796", "34073733364", "34073733355", "queued", "actual visual inspection", "not product-browser acceptance", "#270"):
+            self.assertIn(marker, current)
+
+    def test_historical_rows_cannot_replace_response_observation_checkpoint(self) -> None:
+        text = BASELINE.read_text(encoding="utf-8")
+        current = bounded_section(text, "#### Published response and observation: 01:45 UTC", "#### Published input descendants: 00:40 UTC")
+        for owner in (268, 269):
+            row = active_pr_row(current, owner)
+            for replacement in ("", row.replace("Published; Draft", "Local only; Draft")):
+                mutated = text.replace(row, replacement, 1) + "\nHistorical evidence:\n" + row
+                with patch.object(pathlib.Path, "read_text", return_value=mutated):
+                    with self.assertRaises(AssertionError):
+                        self.test_response_observation_checkpoint_binds_each_published_owner()
+
+    def test_input_descendants_bind_publication_and_coverage_to_each_owner(self) -> None:
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "#### Published input descendants: 00:40 UTC",
+            "### Latest verified cut: 2026-09-06",
+        )
+        for owner, head, parent, coverage in (
+            (265, "e94a2372fe3771f9ddf70291d34fc9a7e4770ec9", "43395711", "1297/13618/17315/1442"),
+            (266, "e3885f69df2cf3899184209efdee5b11bba1bd86", "e94a2372", "1310/13766/17524/1452"),
+            (267, "ebd507ae56c3064e3cae5566502f539c20618a8f", "e3885f69", "1318/13852/17610/1456"),
+        ):
+            row = active_pr_row(current, owner)
+            for marker in ("Published; Draft", parent, coverage):
+                self.assertIn(marker, row)
+            self.assertIn(
+                f"[`{head[:8]}`](https://github.com/ContextualWisdomLab/OriginWeave/commit/{head})",
+                row,
+            )
+        for marker in (
+            "e7fb1527", "009f9a41", "4e020e16", "23 focused", "145 Python",
+            "34068277243", "34068882527", "34070121583", "queued",
+            "zero unresolved review threads", "one counted approval", "seven required workflows",
+            "actual screenshots", "not product-browser acceptance", "own checks and visual inspection",
+            "#268", "protected-main asset preservation", "status receipt",
+        ):
+            self.assertIn(marker, current)
+
+    def test_historical_rows_cannot_hide_current_input_evidence_changes(self) -> None:
+        text = BASELINE.read_text(encoding="utf-8")
+        current = bounded_section(
+            text, "#### Published input descendants: 00:40 UTC",
+            "### Latest verified cut: 2026-09-06",
+        )
+        for owner in (265, 266, 267):
+            row = active_pr_row(current, owner)
+            for replacement in ("", row.replace("Published; Draft", "Local only; Draft")):
+                with self.subTest(owner=owner, replacement=replacement):
+                    mutated = text.replace(row, replacement, 1) + "\nHistorical evidence:\n" + row
+                    self.assertNotEqual(text, mutated)
+                    with patch.object(pathlib.Path, "read_text", return_value=mutated):
+                        with self.assertRaises(AssertionError):
+                            self.test_input_descendants_bind_publication_and_coverage_to_each_owner()
+
+    def test_published_descendants_bind_evidence_to_current_owner_rows(self) -> None:
+        current = bounded_section(
+            BASELINE.read_text(encoding="utf-8"),
+            "#### Published subscription descendants: 23:27 UTC",
+            "#### Connection-bound text responses: 14:54 UTC",
+        )
+        for owner, head, parent, coverage in (
+            (263, "4868d3e9f19133ac3382ee8532878aef27468893", "46ae62aa", "1244/13053/16729/1422"),
+            (264, "433957117ad9e29b26715b062f5adcc9789744ba", "4868d3e9", "1282/13461/17140/1440"),
+        ):
+            row = active_pr_row(current, owner)
+            for marker in ("Published; Draft", head, parent, coverage):
+                self.assertIn(marker, row)
+            self.assertIn(
+                f"[`{head[:8]}`](https://github.com/ContextualWisdomLab/OriginWeave/commit/{head})",
+                row,
+            )
+        for marker in (
+            "zero unresolved review threads", "supersedes the older sole-#147",
+            "12 Ready candidates remain BLOCKED", "one counted approval", "seven required workflows",
+            "13 focused", "145 Python", "actual in-app screenshots", "not product-browser acceptance",
+            "34065055213", "34066516991", "34066516992", "queued",
+            ".github#1929", "protected-main asset preservation", "own checks and visual inspection",
+        ):
+            self.assertIn(marker, current)
+
+    def test_historical_rows_cannot_hide_current_descendant_evidence_changes(self) -> None:
+        text = BASELINE.read_text(encoding="utf-8")
+        current = bounded_section(
+            text,
+            "#### Published subscription descendants: 23:27 UTC",
+            "#### Connection-bound text responses: 14:54 UTC",
+        )
+        for owner in (263, 264):
+            row = active_pr_row(current, owner)
+            for replacement in ("", row.replace("Published; Draft", "Local only; Draft")):
+                with self.subTest(owner=owner, replacement=replacement):
+                    mutated = text.replace(row, replacement, 1) + "\nHistorical evidence:\n" + row
+                    self.assertNotEqual(text, mutated)
+                    with patch.object(pathlib.Path, "read_text", return_value=mutated):
+                        with self.assertRaises(AssertionError):
+                            self.test_published_descendants_bind_evidence_to_current_owner_rows()
+
+    def test_historical_text_cannot_supply_missing_subscription_repair_head(self) -> None:
+        text = BASELINE.read_text(encoding="utf-8")
+        marker = "46ae62aa31e35c702cd61c16322d05c7a9c35da1"
+        mutated = text.replace(marker, "subscription repair head removed", 1)
+        self.assertNotEqual(mutated, text)
+        mutated += "\nHistorical evidence: " + marker
+        with patch.object(pathlib.Path, "read_text", return_value=mutated):
+            with self.assertRaises(AssertionError):
+                self.test_subscription_checkpoint_records_verified_repair_and_visual_gap()
+
+    def test_historical_prose_cannot_hide_latest_root_or_lineage_removal(self) -> None:
+        text = BASELINE.read_text(encoding="utf-8")
+        cases = (
+            (
+                "#195 `63997bcf555e2c5c8e91ba287734ffba3837a1b7`",
+                "foundation evidence removed",
+                self.test_current_snapshot_records_repaired_webdriver_bidi_lineage,
+            ),
+            (
+                "Ready roots: #37, #50,",
+                "Ready roots: #37,",
+                self.test_latest_executable_queue_uses_current_ready_roots,
+            ),
+        )
+        for original, replacement, check in cases:
+            with self.subTest(original=original):
+                mutated = text.replace(original, replacement, 1)
+                self.assertNotEqual(mutated, text)
+                with patch.object(pathlib.Path, "read_text", return_value=mutated):
+                    with self.assertRaises(AssertionError):
+                        check()
+
+    def test_historical_snapshot_preserves_webdriver_bidi_lineage(self) -> None:
+        """The active stack must retain exact heads and the macOS race boundary."""
+        text = BASELINE.read_text(encoding="utf-8")
+        current = text.split("## Current live delivery state", 1)[1].split(
+            "## Observed snapshot: ", 1
+        )[0]
+
+        for marker in (
+            "#195 is Draft at exact head `48eb2d23009c1c804520dd5efcd0d4d072aacef1`",
+            "50 consecutive focused regression passes",
+            "#242 is Draft at exact head `2d0e9f69df9ade21d8e8e3d807c3ff644d83b310`",
+            "PR #243 adopts that corrected parent by ordinary merge at `97fab641ed9d76e6c515eadcef0629edfc8064a3`",
+            "#93 is Draft at exact head `0664f0452cb329cd692cce7f61f9001652abfda2`",
+            "SemanticNodeActionBinding",
+            "does not authorize policy or execute input",
+            "#95 is Draft at exact head `97aa0f2e340ee6fd920d0418f97af276b190554f`",
+            "Only `Decision::Allow`",
+            "`NotAdmitted`",
+            "#96 is Draft at exact head `b7ba8dd1433410cee43084a73e31816da841b2a2`",
+            "callback is never invoked",
+            "#101 is Draft at exact head `bc810b121bb0303f55afa8777a23cc0f9748c1db`",
+            "known-disabled interactive action",
+            "#102 is Draft at exact head `a123c55d4839dae1db7e6671f7d4d158c7cfd9db`",
+            "ObservationAuthorityMismatch",
+            "#103 is Draft at exact head `8b3416169346fa04b53c915b813d55ccf47d1876`",
+            "browser authority first",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, current)
+
+        self.assertIn(
+            "| #195 | Draft | `6922dd98779e8f8aad132a3b1f563d7ba6e6d070` | "
+            "`48eb2d23009c1c804520dd5efcd0d4d072aacef1` |",
+            text,
+        )
+        self.assertIn(
+            "| #93 | Draft | `802ec806cdd4560eab48c484f435766ecabda353` | "
+            "`0664f0452cb329cd692cce7f61f9001652abfda2` |",
+            text,
+        )
+        self.assertIn(
+            "| #242 | Draft | `48eb2d23009c1c804520dd5efcd0d4d072aacef1` | "
+            "`55fef0c3fae1724eddada53e52c4a0311f509aa3` |",
+            text,
+        )
+        self.assertIn(
+            "| #95 | Draft | `0664f0452cb329cd692cce7f61f9001652abfda2` | "
+            "`97aa0f2e340ee6fd920d0418f97af276b190554f` |",
+            text,
+        )
+        self.assertIn(
+            "| #96 | Draft | `97aa0f2e340ee6fd920d0418f97af276b190554f` | "
+            "`b7ba8dd1433410cee43084a73e31816da841b2a2` |",
+            text,
+        )
+        self.assertIn(
+            "| #101 | Draft | `b7ba8dd1433410cee43084a73e31816da841b2a2` | "
+            "`bc810b121bb0303f55afa8777a23cc0f9748c1db` |",
+            text,
+        )
+        self.assertIn(
+            "| #102 | Draft | `bc810b121bb0303f55afa8777a23cc0f9748c1db` | "
+            "`a123c55d4839dae1db7e6671f7d4d158c7cfd9db` |",
+            text,
+        )
+        self.assertIn(
+            "| #103 | Draft | `a123c55d4839dae1db7e6671f7d4d158c7cfd9db` | "
+            "`8b3416169346fa04b53c915b813d55ccf47d1876` |",
+            text,
+        )
+
+    def test_current_snapshot_records_origin_bound_socket_ports(self) -> None:
+        """The live network root must retain its exact port-authority repair."""
+        text = BASELINE.read_text(encoding="utf-8")
+        current = text.split("## Current live delivery state", 1)[1].split(
+            "## Observed snapshot: ", 1
+        )[0]
+
+        for marker in (
+            "PR #50 is Ready at exact head `ad87cfea59db711cb29ef90559790ba77e22029f`",
+            "Fresh independent verification passed all seven fresh-resolution tests and 152 Python contracts",
+            "origin-approved IP could be paired with a different service port",
+            "binds the socket port to the effective scheme-host-port origin",
+            "function/line/region/branch coverage pass locally at 100%",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, current)
+
+    def test_current_snapshot_records_content_length_persistence_repair(self) -> None:
+        """A declared body boundary must not be confused with transport closure."""
+        text = BASELINE.read_text(encoding="utf-8")
+        current = text.split("## Current live delivery state", 1)[1].split(
+            "## Observed snapshot: ", 1
+        )[0]
+
+        for marker in (
+            "PR #37 is Ready at exact head `1e2f41072854edcdbaf0f9ecf14697a3bfd62195`",
+            "returns after the exact declared bytes",
+            "already-buffered surplus remains fail-closed",
+            "does not require TLS EOF",
+        ):
+            with self.subTest(marker=marker):
+                self.assertIn(marker, current)
+
+    def test_changelog_marks_superseded_warc_head_as_historical(self) -> None:
+        """A predecessor WARC head must not look like the current exact evidence."""
+        text = (ROOT / "CHANGELOG.md").read_text(encoding="utf-8")
+
+        self.assertIn("`d83748a70bd1b16dbfec46007fe02989ba6ce188` was superseded", text)
+        self.assertIn("0341079331f9cea669eb9a5cc21842fd6027431e", text)
+
+    def test_baseline_records_central_required_workflow_failures(self) -> None:
+        """The baseline must preserve exact central workflow provenance and failures."""
+        text = BASELINE.read_text(encoding="utf-8")
+
+        for phrase in (
+            "central `.github` repository",
+            "repository ID `1274066402`",
+            "close-empty-pr",
+            "opencode-review",
+            "pr-review-merge-scheduler",
+            "security-scan",
+            "strix",
+            "sast-semgrep",
+            "noema-review",
+            "OpenCode current-head verdict",
+            "Strix provider/backend",
+            "internal server error",
+            "33177641855",
+            "33177641888",
+            "33182772296",
+            "MODEL_OUTPUT_UNAVAILABLE",
+            "model pool exhausted",
+            "33182749298",
+            "#1391",
+            "e4ba6b599cd1e50d0139762885682607b731655d",
+            "did not prove missing workflow identities",
+            "does not authorize bypass",
+        ):
+            with self.subTest(phrase=phrase):
+                self.assertIn(phrase, text)
+
+    def test_same_day_prior_inventory_is_bound_to_the_maintenance_record(self) -> None:
+        """A same-day comparison must retain its exact prior observation in the record."""
+        text = BASELINE.read_text(encoding="utf-8")
+        record_end = "#### Current exact-head active PR evidence"
+        self.assertIn(record_end, text)
+        record = text.split("#### 2026-08-29 maintenance-loop record", 1)[1].split(
+            record_end, 1
+        )[0]
+        self.assertIn("115 open pull requests (31 ready, 84 draft)", record)
+
+    def test_historical_pr_219_head_is_a_full_exact_sha(self) -> None:
+        """The retained PR #219 regression anchor must identify its real commit."""
+        text = BASELINE.read_text(encoding="utf-8")
+
+        self.assertIn(
+            "| #219 | Ready | `0841d2ab3d8b5e60a03c0a8e818cf438e2716829` | "
+            "`8145d40f1b028a8f4dc7e7da47ac89bb9e5bb2c7` |",
+            text,
+        )
+
+    def test_issue_table_distinguishes_open_issues_from_governance_signals(self) -> None:
+        """The table total must distinguish product issues from governance signals."""
+        text = BASELINE.read_text(encoding="utf-8")
+        table_end = "## Buyer-visible and technical gap matrix"
+        self.assertIn(table_end, text)
+        table = text.split("### Open issues and governance signals", 1)[1].split(
+            table_end, 1
+        )[0]
+        self.assertIn("11 open issues plus 2 governance signals", table)
+        self.assertIn("Issue or signal", table)
+
+    def test_evidence_commands_reproduce_inventory_checks_and_review_state(self) -> None:
+        """The baseline links to the executable procedure without duplicating it."""
+        text = BASELINE.read_text(encoding="utf-8")
+        evidence = text.split("## Evidence commands", 1)[1].split("\n## ", 1)[0]
+        script = (ROOT / "scripts" / "ci" / "collect_live_merge_evidence.sh").read_text(
+            encoding="utf-8"
+        )
+
+        self.assertIn("[collect_live_merge_evidence.sh]", evidence)
+        self.assertNotIn("```bash", evidence)
+
+        for phrase in (
+            '"repos/$REPOSITORY/pulls?state=open&per_page=100"',
+            "set -euo pipefail",
+            'EVIDENCE_DIR="$(mktemp -d /tmp/originweave-evidence.XXXXXX)"',
+            '"$EVIDENCE_DIR/open-pr-pages.json"',
+            "jq '[.[][]]' \"$EVIDENCE_DIR/open-pr-pages.json\"",
+            '"repos/$REPOSITORY/issues?state=open&per_page=100"',
+            '"$EVIDENCE_DIR/open-issue-pages.json"',
+            'map(select(has("pull_request") | not))',
+            "open_non_pr_issues",
+            '"repos/$REPOSITORY/pulls/$PR"',
+            '"repos/$REPOSITORY/commits/$HEAD_SHA/check-runs?per_page=100"',
+            '"repos/$REPOSITORY/commits/$HEAD_SHA/statuses?per_page=100"',
+            '"repos/$REPOSITORY/pulls/$PR/reviews?per_page=100"',
+            '"repos/$REPOSITORY/actions/runs?head_sha=$HEAD_SHA&per_page=100"',
+            "check_runs: [$checks[][].check_runs[]?],",
+            "legacy_statuses: [$statuses[][][]?]",
+            "workflow_runs: $exact_pr_base_workflow_runs,",
+            "reviewThreads(first: 100, after: $endCursor)",
+            "rules/branches/main?per_page=100",
+            '"$EVIDENCE_DIR/main-branch-rule-pages.json"',
+            '"$EVIDENCE_DIR/collaborator-pages.json"',
+            '"$EVIDENCE_DIR/collaborators.json"',
+            '"$EVIDENCE_DIR/pr-${PR}-merge-verdict.json.tmp"',
+            '.state == "APPROVED"',
+            ".submitted_at != null",
+            ".commit_id == $head",
+            "group_by(.reviewer)",
+            "required_approving_review_count",
+            "require_last_push_approval",
+            "last_push_approval_authority",
+            '"github_rule_evaluation_required"',
+            "if $require_last_push_approval == true then false",
+            "$pr[0].user.login",
+            '.type == "workflows"',
+            ".parameters.workflows",
+            "required_status_checks",
+            '"$EVIDENCE_DIR/pr-${PR}-merge-verdict.json"',
+            "for ATTEMPT in 1 2 3; do",
+            "RECHECKED_HEAD_SHA=",
+            "RECHECKED_BASE_SHA=",
+            '"$RECHECKED_BASE_SHA" == "$BASE_SHA" &&',
+        ):
+            with self.subTest(phrase=phrase):
+                self.assertIn(phrase, script)
+
+        self.assertNotIn("while :; do", script)
+        self.assertNotIn("/tmp/originweave-open-pr", script)
+        self.assertNotIn("check_runs: [$checks[]?.check_runs[]?],", script)
+        self.assertNotIn("legacy_statuses: [$statuses[][]?]", script)
+        self.assertNotIn("workflow_runs: [$workflow_runs[]?.workflow_runs[]?],", script)
+        self.assertNotIn("$reviews[][]?\n          | select(.state", script)
+        self.assertNotIn("head-commit.json", script)
+        self.assertNotIn("$head_commit[0].committer.login", script)
+        self.assertNotIn("$head_commit[0].author.login", script)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_documentation_active_pr_evidence_contract.py b/tests/test_documentation_active_pr_evidence_contract.py
index a0d2ff3d0..eb9912370 100644
--- a/tests/test_documentation_active_pr_evidence_contract.py
+++ b/tests/test_documentation_active_pr_evidence_contract.py
@@ -1,333 +1,9 @@
-"""Regression contracts for volatile active-PR evidence in canonical documentation."""
+"""Discovery entry point for preserved active-PR baseline history contracts."""
 
-from pathlib import Path
-import re
-import unittest
+from _historical_baseline_contract_loader import export_public, load_historical_contract
 
-
-ROOT = Path(__file__).resolve().parents[1]
-DOCS = ROOT / "docs"
-FITNESS = DOCS / "DOCUMENTATION_FITNESS.md"
-MATURITY = DOCS / "evidence" / "2026-08-10-active-pr-maturity.md"
-BASELINE = DOCS / "product-technical-gap-baseline.md"
-CHANGELOG = ROOT / "CHANGELOG.md"
-
-
-def active_pr_row(text: str, pr_number: int) -> str:
-    """Return exactly one maturity row for an active pull request."""
-    prefix = f"| #{pr_number} |"
-    rows = [line for line in text.splitlines() if line.startswith(prefix)]
-    if len(rows) != 1:
-        raise AssertionError(
-            f"expected exactly one active maturity row for PR #{pr_number}, got {len(rows)}"
-        )
-    return rows[0]
-
-
-def bounded_section(text: str, start: str, end: str) -> str:
-    """Return one explicitly bounded documentation section, failing closed on drift."""
-    if start not in text:
-        raise AssertionError(f"missing section start marker: {start}")
-    remainder = text.split(start, 1)[1]
-    if end not in remainder:
-        raise AssertionError(f"missing section end marker after {start}: {end}")
-    return remainder.split(end, 1)[0]
-
-
-class ActivePullRequestDocumentationContractTests(unittest.TestCase):
-    """Keep volatile implementation evidence separate from protected-main truth."""
-
-    @classmethod
-    def setUpClass(cls) -> None:
-        cls.fitness = FITNESS.read_text(encoding="utf-8")
-        cls.maturity = MATURITY.read_text(encoding="utf-8")
-        cls.baseline = BASELINE.read_text(encoding="utf-8")
-        cls.changelog = CHANGELOG.read_text(encoding="utf-8")
-
-    def test_latest_live_pr_snapshot_is_recorded_in_the_product_baseline(self) -> None:
-        """The current section must preserve exact heads for the newest active slices."""
-        current = bounded_section(
-            self.baseline,
-            "#### Current exact-head active PR evidence",
-            "#### Historical 2026-08-26 maintenance-loop record",
-        )
-        for marker in (
-            "| #73 | Draft | `600d3975c02b68da1974a4c73069b966b39dce7b` | `ce1b138509ab4f52cb0f80290f104358473c6ed3` |",
-            "| #72 | Draft | `f86ce504138e79d6e95141a441f60b40920e1fa6` | `600d3975c02b68da1974a4c73069b966b39dce7b` |",
-            "| #46 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `373113119446d99f578febd39efc19366e7736b1` |",
-            "| #70 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `441a8ce1d09c329c5c1168f4906d9a38fd0abc01` |",
-            "| #82 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `f5776f5f233ac0a7c05e3f4a2846436c23438043` |",
-            "| #152 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `81407a0e5189a413d1be0963fea90a0c2f254ce1` |",
-            "| #210 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `7946dce9a3dd074047d93fca299d48c7aef40e47` |",
-            "| #237 | Draft | `542ca1e9c0a863595b8b6697790005d2471f5413` | `2459af602e72fbfe1ce816919473a1075ec0c41f` |",
-            "| #239 | Draft | `e45cd6cdcdee73b5c16dc942e6c98cb7e745fae0` | `e840ca299d29a15223c8b9bb1397002c4f41b4a3` |",
-            "| #229 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `0145ccba5901e301b41d4be674ca1ed23483ad37` |",
-            "| #220 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `b11db2be68f9b6d71aa4c4290b97a8b22097b353` |",
-            "| #211 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `52a918577958a5701e1146c7eb8b62fe8f8ccd44` |",
-            "| #195 | Draft | `6922dd98779e8f8aad132a3b1f563d7ba6e6d070` | `48eb2d23009c1c804520dd5efcd0d4d072aacef1` |",
-            "| #242 | Draft | `48eb2d23009c1c804520dd5efcd0d4d072aacef1` | `55fef0c3fae1724eddada53e52c4a0311f509aa3` |",
-            "| #93 | Draft | `802ec806cdd4560eab48c484f435766ecabda353` | `0664f0452cb329cd692cce7f61f9001652abfda2` |",
-            "| #95 | Draft | `0664f0452cb329cd692cce7f61f9001652abfda2` | `97aa0f2e340ee6fd920d0418f97af276b190554f` |",
-            "| #96 | Draft | `97aa0f2e340ee6fd920d0418f97af276b190554f` | `b7ba8dd1433410cee43084a73e31816da841b2a2` |",
-            "| #101 | Draft | `b7ba8dd1433410cee43084a73e31816da841b2a2` | `bc810b121bb0303f55afa8777a23cc0f9748c1db` |",
-            "| #102 | Draft | `bc810b121bb0303f55afa8777a23cc0f9748c1db` | `a123c55d4839dae1db7e6671f7d4d158c7cfd9db` |",
-            "| #103 | Draft | `a123c55d4839dae1db7e6671f7d4d158c7cfd9db` | `8b3416169346fa04b53c915b813d55ccf47d1876` |",
-            "| #124 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `fdb88698ca20626a6643bc2ad7944fb968835700` |",
-            "| #37 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | `5e3dfcbd7a4daea297782cb99635990368589232` |",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, current)
-        self.assertNotIn("| #238 |", current)
-        self.assertIn("PR #238 itself", current)
-
-    def test_baseline_refresh_changelog_matches_the_live_snapshot(self) -> None:
-        """The current changelog refresh item must carry its own live queue evidence."""
-        added = self.changelog.split("### Added", 1)[1].split("### Changed", 1)[0]
-        changed = self.changelog.split("### Changed", 1)[1].split("### Security", 1)[0]
-        refresh_prefix = "- Revalidated the product-gap queue at"
-        refresh_lines = [line for line in added.splitlines() if line.startswith(refresh_prefix)]
-        self.assertEqual(1, len(refresh_lines))
-        refresh_line = refresh_lines[0]
-        latest_cut = bounded_section(
-            self.baseline,
-            "### Latest verified cut: 2026-09-09",
-            "### Previous verified cut: 2026-09-08",
-        )
-        current = latest_cut
-        queue_counts = re.findall(
-            r"\*\*(\d+) open pull requests: (\d+) Ready/non-draft and (\d+) Draft; "
-            r"(\d+) open non-PR issues\*\*",
-            " ".join(current.split()),
-        )
-        self.assertEqual(1, len(queue_counts))
-        total, ready, draft, issues = queue_counts[0]
-        inventory_lines = [
-            line for line in self.changelog.splitlines()
-            if line.startswith("- Current delivery inventory:")
-        ]
-        self.assertEqual(1, len(inventory_lines))
-        self.assertIn(f"{total} open pull requests ({ready} ready, {draft} draft)", inventory_lines[0])
-        self.assertIn(f"{issues} open non-PR issues", inventory_lines[0])
-        self.assertIn("Observed 2026-09-09", inventory_lines[0])
-        self.assertIn(
-            "Revalidated the active ruleset inventory at 7 required workflows",
-            changed,
-        )
-        self.assertIn("`codeql-pr`", changed)
-        self.assertIn("Made the open non-PR issue count reproducible", changed)
-        self.assertNotIn("- Corrected the 2026-08-28 product-gap snapshot", added)
-        self.assertNotIn("- Revalidated the product-gap queue at", changed)
-        self.assertNotIn("115 open pull requests (31 ready, 84 draft)", refresh_line)
-        self.assertNotIn("126 open pull requests (54 ready, 72 draft)", refresh_line)
-        self.assertNotIn("128 open pull requests (54 ready, 74 draft)", refresh_line)
-        self.assertNotIn("153 open pull requests (39 ready, 114 draft)", refresh_line)
-
-    def test_latest_inventory_drift_cannot_be_hidden_by_historical_counts(self) -> None:
-        """Changing only the newest count must invalidate an unchanged changelog."""
-        latest = bounded_section(
-            self.baseline,
-            "### Latest verified cut: 2026-09-09",
-            "### Previous verified cut: 2026-09-08",
-        )
-        self.assertIn("131 open pull requests", latest)
-        mutated_latest = latest.replace(
-            "131 open pull requests",
-            "132 open pull requests",
-            1,
-        )
-        self.assertNotEqual(latest, mutated_latest)
-        probe = ActivePullRequestDocumentationContractTests(
-            "test_baseline_refresh_changelog_matches_the_live_snapshot"
-        )
-        probe.baseline = self.baseline.replace(latest, mutated_latest, 1)
-        probe.changelog = self.changelog
-        with self.assertRaises(AssertionError):
-            probe.test_baseline_refresh_changelog_matches_the_live_snapshot()
-
-    def test_current_warc_provider_failures_are_bound_to_current_head(self) -> None:
-        """WARC evidence must describe the current parent head after stack merge."""
-        for marker in (
-            "PR #210 current exact head is `7946dce9a3dd074047d93fca299d48c7aef40e47`",
-            "`Rust contracts` job `98942518975` and `Production coverage` job `98942518680` succeeded",
-            "`noema-review` job `98942513421` and `strix` job `98942803402` remain in progress",
-            "Its predecessor head `5f59947f5e4b0d3bc0aa5b2d4c6722d3b7c43047`, prior stack merge head `66f360ccac5cec60c72222cc79d58e39f6f00088`, earlier exact head `bea65643109449d63d367a35b8d9bf327ee7cb2c`, and their OpenCode/Strix provider failures remain historical evidence only",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, self.baseline)
-
-    def test_current_opencode_dispatch_failures_are_bound_to_current_heads(self) -> None:
-        """OpenCode dispatch authorization failures must remain explicit blockers."""
-        for marker in (
-            "A direct central `opencode-review` dispatch run `33192478312` / job `98921183278` was rejected",
-            "The immediately preceding PR #238 head `d0b0d1ed92f891f14646fc673b8e1c0d912586fd` remains historical",
-            "automatic OpenCode run `33193822920` / job `98926243116` failed closed without a current-head verdict",
-            "central dispatch run `33194506918` / job `98928580387` also failed closed at OpenCode",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, self.baseline)
-
-    def test_dependency_stacks_are_explicit_and_non_shipped(self) -> None:
-        """Current browser, network, sensitive and compatibility stacks stay active-only."""
-        for pr_number in (52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66):
-            with self.subTest(pr_number=pr_number):
-                row = active_pr_row(self.maturity, pr_number)
-                self.assertIn("**IMPLEMENTED_ON_ACTIVE_PR**", row)
-                self.assertNotIn("IMPLEMENTED_ON_PROTECTED_MAIN", row)
-
-        for stack in (
-            "#47 → #50 → #54",
-            "#45 → #46 → #53 → #55",
-            "#40→#52→#57→#58",
-            "#43→#56→#59→#60→#61",
-            "#51→#66",
-        ):
-            with self.subTest(stack=stack):
-                self.assertIn(stack, self.fitness)
-
-        self.assertIn("#49", self.fitness)
-
-    def test_semantic_relationship_evidence_stays_bounded_and_authority_scoped(self) -> None:
-        """PR #52 cannot turn relationship metadata into browser or execution authority."""
-        row = active_pr_row(self.maturity, 52)
-        for marker in ("128", "relationship", "session/context/origin/document"):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, row)
-
-        for marker in (
-            "same browser session, browsing context, canonical origin and document epoch",
-            "Self-parent/self-child relationships and duplicate child handles fail closed",
-            "relationship graph remains descriptive evidence",
-            "not a browser observation adapter",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, self.fitness)
-
-    def test_typed_semantic_query_evidence_stays_descriptive_and_bounded(self) -> None:
-        """PR #57 cannot turn semantic matching into selector or execution authority."""
-        row = active_pr_row(self.maturity, 57)
-        for marker in (
-            "SemanticNodeQuery",
-            "role",
-            "accessible-name",
-            "typed-action",
-            "no CSS/XPath/raw DOM selector language",
-            "browser I/O or action authority",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, row)
-
-        self.assertIn("Draft stacked on exact #52 head", row)
-        self.assertIn("CI run `31429995885`", row)
-        self.assertIn("CodeRabbit exact-head status succeed", row)
-        self.assertIn("remains Draft because #52/#40 are active prerequisites", row)
-
-    def test_sensitive_audience_evidence_does_not_claim_authentication(self) -> None:
-        """An internal audience field is not authenticated workload/service identity."""
-        row = active_pr_row(self.maturity, 55)
-        self.assertIn("authenticated workload/service identity", row)
-        self.assertIn(
-            "audience string accepted by the value/policy primitive is **not authentication**",
-            self.fitness,
-        )
-        self.assertIn("new deployment topology or physical ERD entity", self.fitness)
-
-    def test_mv3_mutation_and_isolation_are_compatibility_not_agent_authority(self) -> None:
-        """Real MV3 evidence must remain separate from OriginWeave capability grants."""
-        bookmark_row = active_pr_row(self.maturity, 56)
-        for marker in ("create", "get", "remove", "compatibility evidence only"):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, bookmark_row)
-
-        for pr_number in (59, 60, 61):
-            with self.subTest(pr_number=pr_number):
-                row = active_pr_row(self.maturity, pr_number)
-                self.assertIn("**IMPLEMENTED_ON_ACTIVE_PR**", row)
-                self.assertNotIn("IMPLEMENTED_ON_PROTECTED_MAIN", row)
-
-        self.assertIn("Manifest V3 compatibility", self.fitness)
-        self.assertIn(
-            "Chromium permission or browser compatibility success is not an OriginWeave Agent capability",
-            self.fitness,
-        )
-        self.assertIn(
-            "#43/#49/#56/#59/#60/#61 are active compatibility evidence only",
-            self.fitness,
-        )
-        self.assertIn("Update migration is intentionally distinct from restart persistence", self.fitness)
-        self.assertIn("isolated-world behavior is intentionally distinct from injection alone", self.fitness)
-
-    def test_extension_proposal_grant_does_not_become_agent_policy_authority(self) -> None:
-        """PR #62 must remain a policy-isolation regression, not a new action grant."""
-        row = active_pr_row(self.maturity, 62)
-        for marker in (
-            "ProposeTypedAction",
-            "out-of-grant target origin",
-            "missing core `Navigate` capability",
-            "untrusted instruction source",
-            "adds no production API or real Chromium adapter",
-            "does not convert extension proposal authority into Agent action/origin authority",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, row)
-        self.assertIn("CI run `31436844685`", row)
-        self.assertIn("Security Scan run `31436844615`", row)
-        self.assertIn("SAST Semgrep run `31436844646`", row)
-        self.assertNotIn("IMPLEMENTED_ON_PROTECTED_MAIN", row)
-
-    def test_latest_agent_task_and_secret_composition_evidence_remains_partial(self) -> None:
-        """Newest active slices must not be promoted into a complete browser or broker runtime."""
-        secret_approval = active_pr_row(self.maturity, 63)
-        for marker in (
-            "ProposeTypedAction",
-            "RequireApproval(RiskClass::R3)",
-            "no secret broker",
-        ):
-            with self.subTest(pr_number=63, marker=marker):
-                self.assertIn(marker, secret_approval)
-
-        action_outcome = active_pr_row(self.maturity, 64)
-        for marker in (
-            "PostConditionPredatesDispatch",
-            "monotonic",
-            "not a browser dispatcher",
-        ):
-            with self.subTest(pr_number=64, marker=marker):
-                self.assertIn(marker, action_outcome)
-
-        controlled_fixture = active_pr_row(self.maturity, 65)
-        for marker in (
-            "controlled",
-            "prompt-injection",
-            "not a browser adapter",
-        ):
-            with self.subTest(pr_number=65, marker=marker):
-                self.assertIn(marker, controlled_fixture)
-
-        process_set = active_pr_row(self.maturity, 66)
-        for marker in (
-            "process-set RSS",
-            "duplicate",
-            "does not discover Chromium PIDs",
-        ):
-            with self.subTest(pr_number=66, marker=marker):
-                self.assertIn(marker, process_set)
-
-        for marker in (
-            "#62/#63",
-            "#64",
-            "#65",
-            "#51→#66",
-            "real Chromium",
-        ):
-            with self.subTest(fitness_marker=marker):
-                self.assertIn(marker, self.fitness)
-
-    def test_erd_stays_conceptual_without_persistence_owner(self) -> None:
-        """Active in-memory/value primitives must not manufacture a physical data model."""
-        self.assertIn("Conceptual ERD/domain model", self.fitness)
-        self.assertIn("add no OriginWeave-owned durable store", self.fitness)
-        self.assertIn("false architecture", self.fitness)
-
-
-if __name__ == "__main__":
-    unittest.main()
+_legacy = load_historical_contract(
+    "legacy_documentation_active_pr_evidence_contract.py",
+    "_originweave_legacy_documentation_active_pr_evidence_contract",
+)
+export_public(_legacy, globals())
diff --git a/tests/test_gap_snapshot_inventory_consistency.py b/tests/test_gap_snapshot_inventory_consistency.py
index d4c9c0599..c01154c45 100644
--- a/tests/test_gap_snapshot_inventory_consistency.py
+++ b/tests/test_gap_snapshot_inventory_consistency.py
@@ -1,461 +1,9 @@
-"""Regression contracts for current and historical product-gap inventory evidence."""
+"""Discovery entry point for preserved gap-snapshot history contracts."""
 
-from __future__ import annotations
+from _historical_baseline_contract_loader import export_public, load_historical_contract
 
-from pathlib import Path
-import re
-import unittest
-
-
-ROOT = Path(__file__).resolve().parents[1]
-BASELINE = ROOT / "docs" / "product-technical-gap-baseline.md"
-CHANGELOG = ROOT / "CHANGELOG.md"
-
-
-class GapSnapshotInventoryConsistencyTests(unittest.TestCase):
-    """Keep volatile live truth separate from immutable dated snapshots."""
-
-    @classmethod
-    def setUpClass(cls) -> None:
-        cls.baseline = BASELINE.read_text(encoding="utf-8")
-        cls.changelog = CHANGELOG.read_text(encoding="utf-8")
-
-    def test_current_live_state_is_distinct_from_dated_snapshot(self) -> None:
-        """The live section must bind fresh evidence without promoting queued work."""
-        current = self.baseline.split("## Current live delivery state", 1)[1].split(
-            "## Observed snapshot: 2026-08-29", 1
-        )[0]
-        for marker in (
-            "125 open pull requests",
-            "12 Ready/non-draft",
-            "113 Draft",
-            "13 open non-PR issues",
-            "87c4daa1830bac5a5228b6036752ad5633232085",
-            "18156473",
-            "7 central required workflows",
-            "codeql-pr",
-            "Live GitHub PR/base/head/check APIs are authoritative over PR bodies",
-            "Ready roots are #37 `1e2f41072854edcdbaf0f9ecf14697a3bfd62195`, #50 `ad87cfea59db711cb29ef90559790ba77e22029f`, #166 `e84a1a2cc82b1c666218efd441da97849f47b8c2`, #219 `65e4315d80137badc0b55e1b9617015beb1db568`, #220 `e545b94e1de499b96b867694f80ac04ad247becd`, #229 `024f63690cf05cfe6f0d4a430f0e18ea8fd2c4d6`, #240 `24930a3a9ee79c0b712ee3df6589b0592eb6e18f`, #272 `b1cae8ad1cbd8eb6992037c830aea30b9aa436b3`, #274 `802d0bdff7536d9ac253305d3e0237b4e4a1789e`, #285 `f455c2cd64b3dd3f027c91d396103792a205ddd0`, and #287 `af83c40dd2990a03064a92ca75430a9cc400f098`",
-            "PR #290 is Draft at exact head `ebeefcd534db4324498fdb18046ebc6255ddcdf2`",
-            "PR #238's moving exact head is intentionally omitted",
-            "PR #166 exact head `e84a1a2cc82b1c666218efd441da97849f47b8c2`",
-            "PR #220 exact head `e545b94e1de499b96b867694f80ac04ad247becd`",
-            "retain formal `CHANGES_REQUESTED` decisions",
-            "thread-resolution state could not be refreshed",
-            "115 queued workflow runs",
-            "rerunning unchanged heads would only add duplicate queue load",
-            "PR #249 is Draft at exact head `84b9407978ae0f6c115f01170b6069c601b21104`",
-            "989 functions, 9978 lines, 12717 regions, 1098 branches",
-            "PR #250 now adopts #249 at `ec433b844a121f8554c062f92267991af9cacb6f`",
-            "1023 functions, 10518 lines, 13525 regions, 1186 branches",
-            "PR #251 now adopts #250 at `f02af6d0dd01708d495cc08dec785675f3d58898`",
-            "1037 functions, 10647 lines, 13681 regions, 1192 branches",
-            "PR #252 now adopts #251 at `6569bf40b6595ac74c2f0a997d202137f07ba1db`",
-            "1044 functions, 10699 lines, 13751 regions, 1192 branches",
-            "PR #253 now adopts #252 at `afb623e4449b7cbf926fdcef7225ceaca822cfcf`",
-            "1054 functions, 10747 lines, 13787 regions, 1194 branches",
-            "Review `5120013340` remains unresolved",
-            "PR #254 now adopts #253 at `b11b6c9bccd8335b58a4fb599f8ad29ac419637f`",
-            "1061 functions, 10802 lines, 13843 regions, 1200 branches",
-            "Review `5120077272` remains actionable",
-            "Current #148 `bded4fc9d32e5e047cea99d182aa05ae2cd03bf6`",
-            "all 247 Python contracts pass on Linux with zero skips",
-            "Current #150 `d3c29359fd10e540ca9b3b2723bde6f94866cdb6`",
-            "all 229 Python contracts pass on macOS and Linux",
-            "#166 Strix run `33929688857` / job `101237371800` failed",
-            "console classifier also matches ordinary denied-policy prose",
-            "#219 Strix run `33925442296` / job `101226800608`",
-            "six-hour GitHub job execution limit",
-            "#240 Strix run `33925596890` / job `101227893828`",
-            "Current #147 `3dff28d9bf2dd27b72507e39979d51b8bf140fb4`",
-            "all six focused and all 228 native tests now pass",
-            "all 228 native tests execute but two forced-close protocol-fault subcases fail",
-            "Temporary integration PR #291 is now closed",
-            "PR #142 at `015025f2539e4fb1dbd7d259ec22dad50f944396`",
-            "187 Python contracts passed",
-            "PR #143 at `44fd9a450f864feff5cf2ba2883425a71ba10b9b`",
-            "192 Python contracts",
-            "415 functions, 3555 lines, 4444 regions, 476 branches",
-            "process-observation errors leave termination unproven",
-            "PR #144 at `09f2e087d0c20fe81386c18099e739a9e611a8ad`",
-            "194 Python contracts passed",
-            "separate root and process-set deadline budgets",
-            "Current PR #144 parent adoption is `3c0c5c363c2da0b4e8a621226e2f7766c735b743`",
-            "all 196 Python contracts",
-            "PR #145 now adopts #144 at `bb5e8f834c37f9ce35f84db8ed1146da3659d6aa`",
-            "all 202 Python contracts",
-            "PR #146 review repair is `812c0020bd2ecdb3eda39d999ed3327647550bdf`",
-            "10 assertion failures and 7 uncaught protocol-error cases",
-            "all 224 Python contracts",
-            "concurrent parent-adoption `11944410450684809ee1a71a35c77abafc5358db`",
-            "Noema job `101245089068` failed with HTTP 502 after 2739.2 seconds",
-            "Strix scan job `101243872506` executed successfully",
-            "#250 `0eab23d5e388c5c8b984c0021a58316680c9ba8b`",
-            "#251 `86e8ad76838f2a64aa7e0cd56ba1f931c8d0c3dc`",
-            "#252 `2015259529ada99af836989079cc85a15779a2d8`",
-            "#253 `0d72082e595c0e1fcc03d609ba337896ed14e2fc`",
-            "three unauthenticated caller booleans mint `OperationallyComplete`",
-            "raw teardown claims must remain explicitly unverified",
-            "#254 `cbaf50dcc97753cc73135497ea8225e8b18de190`",
-            "#255 `a13de5f9321e72c1867974eb7a43230f031e58df`",
-            "Follow-up review `5120203295` at #255 head `92392b2da33bcd5446c6d2eb1b5504c39e60e3d6`",
-            "received-message provenance must be checked before correlation is consumed",
-            "Received-connection verification at #255 head `e7bfec4488b7cb4776df7b546cacb46c8c9eb13e`",
-            "12 focused tests passed",
-            "lines=11046/11049 and regions=14067/14071",
-            "Follow-up quality verification at #255 head `63cbca0a98cf9496af981819d98029e656fc4342`",
-            "functions=1082/1083, lines=11046/11051, regions=14071/14074",
-            "unused private accessor",
-            "Prior #255 source-owner repair is `ebac126d1632c94775c2454423575275eec45def`",
-            "PR #255 now adopts #254 at `3e7057443d7c9532ff526acb5eefe8cd4778c767`",
-            "PR #256 now adopts #255 at `ced4a851ca66c08d895a098725c7f0ad3ecf0c38`",
-            "PR #257 now adopts #256 at `f4a8f2cbf515bea348f500b59615a9581c1b96a2`",
-            "1082 functions, 11032 lines, 14086 regions, 1202 branches",
-            "1088 functions, 11077 lines, 14146 regions, 1210 branches",
-            "1093 functions, 11131 lines, 14202 regions, 1214 branches",
-            "1100 functions, 11185 lines, 14272 regions, 1214 branches",
-            "1140 functions, 11841 lines, 15129 regions, 1332 branches",
-            "1154 functions, 11969 lines, 15314 regions, 1334 branches",
-            "Central Strix #1563 advanced to `13fbb48e0b3eeca4ce7d9678add934f9bd87ad3f`",
-            "independently inspected four-file source diff",
-            "2890 passed, 1 skipped, and 21 subtests are owner-reported",
-            "521 functions, 4420 lines, 5349 regions, 654 branches",
-            "#37 Strix run `33946242342`",
-            "Strix run timed out after 900s",
-            "#148 hosted compatibility run `33962062608`",
-            "browser-crash recovery fails 0/3 trials",
-            "#261 now includes actual #260",
-            "1173 functions, 12104 lines, 15511 regions, 1334 branches",
-            "1221 functions, 12781 lines, 16404 regions, 1418 branches",
-            "no subscription bytes but two outstanding commands",
-            "GraphQL rejected both the inventory query and a minimal probe",
-            "REST reviews, check runs, and combined statuses were read for all 125 open PRs",
-            "warning: --branch option is unstable",
-            "numerical coverage enforcement is distinct from warning-free instrumentation",
-            "PR #139 at `b7ea5bfe336456fb263dd479a60b1cd0193d8a47`",
-            "Opening-exchange fixture repair at #242",
-            "PR #243 adopts that corrected parent by ordinary merge",
-            "PR #246 adopts #243 at `585791f3641fbe757c3bd9fd36d5316adcc78d63`",
-            "PR #248 is Draft at `b386f17c4826adabebda084bff2fba35aee94dd0`",
-            "native unittest discovery collected zero checks",
-            "975 functions, 9848 lines, 12563 regions, 1092 branches",
-            "962 functions, 9701 lines, 12427 regions, 1084 branches",
-            "PR #141 at `fbdf64f5818ce0c53b475196d5bcfa2ac9900846`",
-            "183 Python contracts passed",
-            "two informational threads were resolved without source changes",
-            "Noema attempt 2 is queue-admission evidence, not provider recovery or a review verdict",
-            "central CodeQL dispatches remain queued and were not duplicated",
-            "#256 `9f2e6f29be46371762e3031a97c1cac04720694f`",
-            "#257 `ea2b5b78868917219c46f1304558b92490a7f6fe`",
-            "Issue #279",
-            "Issue #28 remains the P0 governed-browser integration target",
-            "PR #260 is Draft at exact head `2c5049aff97a90958e8262b1d403bdcbd64a1e8b`",
-            "PR #258 is Draft at exact head `5f830324f6d5a47ac213a57528ef95649bdfd0df`",
-            "PR #259 is Draft at exact head `66731982ed51ad62a04fcfbc759b0a33721a0254`",
-            "PR #261 is Draft at exact head `934eb7d37568b439c442ffe1d1f6a9c8f8ed58a0`",
-            "Repair PR #277 is Draft at exact head `117f6414e8a6db46eb2b32f4ebae85cf2a208371`",
-            "PR #70 is Draft at exact head `77eb0f2ee71783e06171784b7173c0b4cd530e61`",
-            "DDD/MCP repair #272 is Ready at exact head `b1cae8ad1cbd8eb6992037c830aea30b9aa436b3`",
-            "PR #229 is Ready at exact head `024f63690cf05cfe6f0d4a430f0e18ea8fd2c4d6`",
-            "PR #281 remains Draft at exact head `adaca6427d68f550b39293a69b7c733430d1c385`",
-            "PR #282 is Draft at exact head `b64e0708584beff3fb54acf226cb3e667773e473`",
-            "PR #283 is Draft at exact head `c904300a6a1bda83af24f84d586f1c5f6a6491aa`",
-            "stacked on predecessor #282 head `b54a5856d8201911f05d69622f0d5594a371adf0` rather than current #282 exact `b64e0708584beff3fb54acf226cb3e667773e473`",
-            "must adopt the corrected current parent before its compare can become current evidence",
-            "PR #288 is Draft at exact head `0f434bc29d468127412366b6864b733b40b83c4d`",
-            "all 173 Python contracts pass on that unchanged head",
-            "PR #269 is Draft at exact head `7854394266d3f292e779193c01413a34f6798d7c`",
-            "PR #270 is Draft at exact head `191a14535219ea8033777fa4c970efb281b62418`",
-            "PR #271 is Draft at exact head `802ec806cdd4560eab48c484f435766ecabda353`",
-            "PR #285 is Ready at exact head `f455c2cd64b3dd3f027c91d396103792a205ddd0`",
-            "CI `33930234387`",
-            "Rust contracts `101207153048` and Production coverage `101207153244` succeeded",
-            "native CI success does not replace the seven required central workflows",
-            "Security Scan `33924016851`",
-            "SAST Semgrep `33924016903`",
-            "CodeQL PR `33924016883`",
-            "did not materialize fresh central required-workflow runs",
-            "PR #287 is Ready at exact head `af83c40dd2990a03064a92ca75430a9cc400f098`",
-            "#287 CodeQL handoff RCA",
-            "33954721186",
-            "33955024697",
-            "33955164029",
-            "PR #273 is Draft at exact head `e5c8fcb66bf644dfa750bb1b40ba3d600cb7805a`",
-            "GitHub Releases is empty",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, current)
-        self.assertNotRegex(
-            current,
-            re.compile(r"PR #238 exact head `[0-9a-f]{40}`"),
-            "The self-referential baseline cannot pin the SHA produced by its own commit; read #238 live metadata instead.",
-        )
-        self.assertRegex(
-            current,
-            re.compile(r"Observed at \(UTC\): `\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}Z`"),
-        )
-        for non_passing in (
-            "queued reviewer evidence is non-passing",
-            "discard queued, skipped, cancelled, absent, predecessor, synthetic, status-only, and model-only evidence",
-        ):
-            with self.subTest(non_passing=non_passing):
-                self.assertIn(non_passing.casefold(), current.casefold())
-        self.assertIn("binds Git rename/copy similarity to blob identity", current)
-        self.assertNotIn("remaining rename/copy similarity-binding RED", self.changelog)
-        for stale in (
-            "Protected `main` is `c789b802fc98a8d7fd8c09d9327f36828054d2a1` through #280",
-            "143 open pull requests",
-            "142 open pull requests",
-            "24 Ready/non-draft",
-            "138 Draft",
-            "121 Draft",
-            "12 open non-PR issues",
-            "10 central required workflows",
-            "6 central required workflows",
-            "PR #260 is Draft at exact head `a3741389fdc491c7ecccc20f77609c55bc56d20f`",
-            "PR #261 remains stale-parent Draft at exact head `127e02503e48938e29a9a07410574c7e72fc661a`",
-            "DDD/MCP repair #272 is Draft at exact head `80272f18422c9946077ad9bd674f603db8f020da`",
-            "PR #229 is Draft at exact head `7ae426e760e8351ee792ce9df4266d7e7483d0d4`",
-            "PR #282 is Draft at exact head `b2a120f892973e76c0ea0f06e7105bdf7a268009`",
-            "PR #283 is Draft at exact head `a9603170e848a7c029531fe75727f02992ade2de`",
-            "PR #282 is Draft at exact head `e3a40a4f78a3fbfc751ab9efad321b8207fb43e5`",
-            "PR #283 is Draft at exact head `85d31596c8b7251135f773b1d54d0f656fa10bbf`",
-            "PR #260 is Draft at exact head `56600a6fd982cfafd784f4b7bb659d918113ca90`",
-            "PR #281 remains Draft at exact head `3ed5d7e8cf77547c96feff2cfb24c46d74a73ebb`",
-        ):
-            with self.subTest(stale=stale):
-                self.assertNotIn(stale, current)
-
-    def test_latest_cut_does_not_promote_prior_observations(self) -> None:
-        """Fresh queue evidence must precede the explicitly historical cut."""
-        latest = self.baseline.split("## Current live delivery state", 1)[1].split(
-            "### Prior observation cut: 2026-09-05", 1
-        )[0]
-        for marker in (
-            "Latest verified cut: 2026-09-06",
-            "63997bcf555e2c5c8e91ba287734ffba3837a1b7",
-            "fd589cd693946ef1ce2c9270c2dfb6a1087bdfb9",
-            "0135984f1bc1f68d89d7777f49c4999474105a12",
-            "973e34bc24ae9bdd96a50764f2be6c8603eff66c",
-            "3f22de94b63da83eaa8b5b1270912b21a3ecd006",
-            "GraphQL thread resolution was refreshed for all 125 open PRs",
-            "#147 retains the sole unresolved thread",
-            "34013251657",
-            "33990522248",
-            "34009256997",
-            "#279 owns current-workflow reconstruction",
-            "#212 owns sandbox-helper adoption",
-            "earlier local coverage counts do not validate these newer heads",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, latest)
-
-    def test_current_cut_records_parent_adoption_and_sent_command_repair(self) -> None:
-        """New source evidence must not inherit earlier native or browser acceptance."""
-        latest = self.baseline.split("## Current live delivery state", 1)[1].split(
-            "### Prior observation cut: 2026-09-05", 1
-        )[0]
-        for marker in (
-            "6ab7fc9166f397f49442243492ae88d6ccad56cf",
-            "43d3b5a3a2b5ce4f51a93d1152a0ee82620f4f3e",
-            "73f11de2232060ac7680e188db88ef7609123296",
-            "253 Python contracts",
-            "1273/13294/16963/1430",
-            "34020055803",
-            "actual-resend freshness",
-            "current caller-supplied registry",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, latest)
-
-    def test_current_cut_separates_browser_failure_and_unpublished_repair(self) -> None:
-        """Hosted browser failure and private source proof must stay distinct."""
-        latest = self.baseline.split("## Current live delivery state", 1)[1].split(
-            "### Prior observation cut: 2026-09-05", 1
-        )[0]
-        for marker in (
-            "9985228944",
-            "05b6d1cd6092a9e2e18a5003a5a1faeb73b8a8d9b4fd7533f7f24bc938833436",
-            "5557994915",
-            "bd29f405a6c927b2f7d1c437dccbfb8bcec424f1",
-            "1273/13311/16973/1432",
-            "unpublished checkpoint",
-            "5557964460",
-            "writer acknowledgement",
-            "8ebcc6131a5dd6bf0b4720c2f1ff8d40d1cc39f8",
-            "lines 13327/13330 and regions 17004/17008",
-            "functions 1274/1274 and branches 1434/1434",
-            "f352e3f04bdc3ed1912be7c452c5c426d9caaec7dd1a613b7d3ab9c6486ab69e",
-            "formatting check fails in five test/support files",
-            "Neither private nor remote results establish combined-source acceptance",
-            "2a9fdc5418b9af10353cdad0f6f6470655bf457d",
-            "52cff96954c3fec7b8cda5409e4560e970bfcbdf",
-            "1273/13317/16984/1432",
-            "f7e71e6fd67723688535053c389b4ff718c4563b403924af462848d3d51b541f",
-            "34024499232",
-            "5558261278",
-            "both complete lineages are now ancestors",
-            "too_many_active_automations",
-            "10f138f8787d596e8b556fe50c9e4e52bc1295b7",
-            "b3ffeac9",
-            "1278/13371/17056/1434",
-            "9f5b4942d6a040a83f09593a2b81406c44bc9d2b402fa995a95294ce8e9cbb7e",
-            "34026519860",
-            "34026519878",
-            "owner destruction",
-            "registry-to-transport association",
-            "805051527cf95e14ba126c9dd3159db86d190224",
-            "1279/13409/17087/1436",
-            "64a62b1e03ee3ed3d62654a3227495e82b7ff93357038ae146d8584c581ac060",
-            "34029687813",
-            "34029687816",
-            "terminal-success predecessor evidence",
-            "actual Edge visual inspection",
-            "GraphQL access recovered",
-            "7fb93e77b800f27187a5c02333298cc31a025bd6",
-            "1279/13421/17106/1438",
-            "277c5bc0a17f16966a8eb388d54e4e172be36356d7dead5bbead9ed53ad8cff6",
-            "34031281825",
-            "34031281812",
-            "new-head visual inspection remains incomplete",
-            "Mac is locked",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, latest)
-        self.assertNotIn(
-            "Native CI `34018588292` and MV3 `34018588298` remain queued",
-            latest,
-        )
-
-    def test_pointer_integration_cut_separates_parent_and_hosted_acceptance(self) -> None:
-        """Published pointer proof must retain the newer parent and visual evidence gaps."""
-        latest = self.baseline.split("#### Current pointer integration and independently owned parent repair", 1)[1].split(
-            "#### Scheduling and central-owner follow-up", 1
-        )[0]
-        for marker in (
-            "35555d0f8491d4ee95c2e61d1a2aaa2c02a0635c",
-            "3f9cdc1a",
-            "1293/13567/17273/1440",
-            "b1374f65538fb6557b4e5d877aa067e183202ed1f56c75be047b3ff2dfdc92a0",
-            "34032661775",
-            "34032661781",
-            "b4702cd503fa3f721e0d1f44b355563753dac0a2",
-            "5559076006",
-            "not latest-parent acceptance",
-            "four silently deleted parent postcondition tests",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, latest)
-        self.assertIn("Browser access recovered", latest)
-        self.assertIn("published body and Checks page", latest)
-        self.assertIn("new documentation head still requires its own visual inspection", latest)
-        self.assertIn("5559140277", latest)
-
-    def test_session_adoption_cut_records_executed_and_visual_evidence(self) -> None:
-        """Current source evidence must not inherit earlier pending-adoption claims."""
-        latest = self.baseline.split("### Latest verified cut: 2026-09-06", 1)[1].split(
-            "#### Prior observation: 12:28 UTC", 1
-        )[0]
-        for marker in (
-            "7147893c96ca95c9b5b275d8011c5bfe99aab065",
-            "6f331a5b220349a1aaa1b1841d5e8ec027b9ad49",
-            "98621adf", "1296/13601/17303/1442",
-            "480d6d3ca492565e8a04bfce4b0c135dffbff276532d129c410674838e8fec2f",
-            "34036335342", "34036335505", "queued",
-            "actual Edge visual inspection", "not product-browser acceptance",
-            "5559497861", "not protected-main delivery",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, latest)
-
-    def test_current_baseline_inventory_matches_the_verified_snapshot(self) -> None:
-        """The dated 2026-08-29 snapshot must keep its exact historical inventory."""
-        current = self.baseline.split("### Open pull requests", 1)[1].split(
-            "#### 2026-08-29 maintenance-loop record", 1
-        )[0]
-        for marker in (
-            "108 open pull requests",
-            "24 non-draft",
-            "84 draft",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, current)
-
-        for stale in (
-            "128 open pull requests",
-            "126 open pull requests",
-            "54 non-draft",
-            "72 draft",
-            "115 open pull requests",
-            "31 non-draft",
-            "116 open pull requests",
-            "32 non-draft",
-            "111 open pull requests",
-            "27 non-draft",
-            "153 open pull requests",
-            "114 draft PRs",
-        ):
-            with self.subTest(stale=stale):
-                self.assertNotIn(stale, current)
-
-    def test_unreleased_changelog_uses_one_current_inventory(self) -> None:
-        """The current Unreleased entry must match the verified volatile inventory."""
-        unreleased = self.changelog.split("## [Unreleased]", 1)[1]
-        preamble, remainder = unreleased.split("### Added", 1)
-        added = remainder.split("### Changed", 1)[0]
-
-        expected = "125 open pull requests (12 ready, 113 draft)"
-        self.assertIn(expected, preamble)
-        self.assertIn("13 open non-PR issues", preamble)
-        self.assertIn(expected, added)
-        self.assertIn("13 open non-PR issues", added)
-        self.assertNotIn("143 open pull requests", preamble)
-        self.assertNotIn("108 open pull requests (24 ready, 84 draft)", preamble)
-
-    def test_current_snapshot_records_recent_stack_merges_and_revalidation(self) -> None:
-        """A stacked merge must update the dated queue and parent exact-head evidence."""
-        current = self.baseline.split("### Open pull requests", 1)[1].split(
-            "#### 2026-08-29 maintenance-loop record", 1
-        )[0]
-        record = self.baseline.split(
-            "#### 2026-08-29 maintenance-loop record", 1
-        )[1].split("#### Current exact-head active PR evidence", 1)[0]
-        self.assertIn("108 open pull requests", current)
-        self.assertIn("24 non-draft", current)
-        self.assertIn("#217 was squash-merged", record)
-        self.assertIn(
-            "#53 was exact head `4ecc81e59ae7bc3a640e65e2442bf30c079bd94c`",
-            record,
-        )
-        self.assertIn(
-            "#217 was exact head `6b8a3fdeae52ad94b90086bbc9b42863b90c9614`",
-            record,
-        )
-        self.assertIn("66f360ccac5cec60c72222cc79d58e39f6f00088", record)
-        self.assertIn("#67 was squash-merged", record)
-        self.assertIn("5021d142583cb5a8e393248048bb824762a98056", record)
-        self.assertIn("PR #64 consequently advanced", record)
-        self.assertIn(
-            "| #210 | Ready | `542ca1e9c0a863595b8b6697790005d2471f5413` | "
-            "`7946dce9a3dd074047d93fca299d48c7aef40e47` |",
-            self.baseline,
-        )
-        self.assertNotIn("| #217 |", current)
-
-    def test_current_documentation_head_records_security_and_review_state(self) -> None:
-        """The dated self-referential record must preserve its historical gate truth."""
-        record = self.baseline.split(
-            "#### 2026-08-29 maintenance-loop record", 1
-        )[1].split("#### Current exact-head active PR evidence", 1)[0]
-        for marker in (
-            "The immediately preceding PR #238 head `d0b0d1ed92f891f14646fc673b8e1c0d912586fd` remains historical",
-            "automatic OpenCode run `33193822920` / job `98926243116` failed closed",
-            "current Strix run `33193822929` / job `98925769697` succeeded",
-            "central dispatch run `33194506918` / job `98928580387` also failed closed at OpenCode",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, record)
-
-
-if __name__ == "__main__":
-    unittest.main()
+_legacy = load_historical_contract(
+    "legacy_gap_snapshot_inventory_consistency.py",
+    "_originweave_legacy_gap_snapshot_inventory_consistency",
+)
+export_public(_legacy, globals())
diff --git a/tests/test_live_gap_evidence_integrity_contract.py b/tests/test_live_gap_evidence_integrity_contract.py
index 998be0669..6b14b49ea 100644
--- a/tests/test_live_gap_evidence_integrity_contract.py
+++ b/tests/test_live_gap_evidence_integrity_contract.py
@@ -1,171 +1,9 @@
-"""Fail-closed contracts for the current product-gap evidence procedure."""
+"""Discovery entry point for preserved live-gap evidence-integrity contracts."""
 
-from pathlib import Path
-import subprocess
-import unittest
+from _historical_baseline_contract_loader import export_public, load_historical_contract
 
-
-ROOT = Path(__file__).resolve().parents[1]
-BASELINE = ROOT / "docs" / "product-technical-gap-baseline.md"
-CHANGELOG = ROOT / "CHANGELOG.md"
-AGENTS = ROOT / "AGENTS.md"
-EVIDENCE_SCRIPT = ROOT / "scripts" / "ci" / "collect_live_merge_evidence.sh"
-
-
-def bounded(text: str, start: str, end: str) -> str:
-    """Return one current section without accepting a historical substitute."""
-    if start not in text:
-        raise AssertionError(f"missing start marker: {start}")
-    remainder = text.split(start, 1)[1]
-    if end not in remainder:
-        raise AssertionError(f"missing end marker after {start}: {end}")
-    return remainder.split(end, 1)[0]
-
-
-class LiveGapEvidenceIntegrityContractTests(unittest.TestCase):
-    """Keep merge evidence bound to the exact current PR state."""
-
-    @classmethod
-    def setUpClass(cls) -> None:
-        cls.baseline = BASELINE.read_text(encoding="utf-8")
-        cls.changelog = CHANGELOG.read_text(encoding="utf-8")
-        cls.agents = AGENTS.read_text(encoding="utf-8")
-        cls.evidence = EVIDENCE_SCRIPT.read_text(encoding="utf-8")
-        cls.latest = bounded(
-            cls.baseline,
-            "### Latest verified cut: 2026-09-09",
-            "### Previous verified cut: 2026-09-08",
-        )
-
-    def test_latest_inventory_and_changelog_use_the_september_9_cut(self) -> None:
-        marker = (
-            "131 open pull requests: 14 Ready/non-draft and "
-            "117 Draft; 14 open non-PR issues"
-        )
-        self.assertIn(marker, " ".join(self.latest.split()))
-        inventory = [
-            line
-            for line in self.changelog.splitlines()
-            if line.startswith("- Current delivery inventory:")
-        ]
-        self.assertEqual(1, len(inventory))
-        self.assertIn("131 open pull requests (14 ready, 117 draft)", inventory[0])
-        self.assertIn("14 open non-PR issues", inventory[0])
-        self.assertIn("Observed 2026-09-09", inventory[0])
-
-    def test_presentation_snapshot_uses_full_exact_sha(self) -> None:
-        full_sha = "6855e2578ae94279cc9ab4a14527b016e8c049ee"
-        self.assertIn(full_sha, self.latest)
-        self.assertNotIn("to exact head\n`6855e257`", self.latest)
-
-    def test_active_presentation_harness_is_not_described_as_unimplemented(self) -> None:
-        normalized = " ".join(self.latest.split())
-        self.assertIn("active controlled evidence harness is implemented", normalized)
-        self.assertIn("failed 0/3 at session creation before navigation", normalized)
-        self.assertIn("product Browser Session adapter", normalized)
-
-    def test_presentation_successor_records_its_repaired_contract_boundary(self) -> None:
-        """Keep the active child repair distinct from browser-runtime acceptance."""
-        normalized = " ".join(self.latest.split())
-        self.assertIn("Draft successor #298", normalized)
-        self.assertIn("d01f45c2c8ac7b0fc4dbc3d3ada60238732cdf8c", normalized)
-        self.assertIn("#305", normalized)
-        self.assertIn("does not establish live Chromium transport", normalized)
-
-    def test_documented_current_evidence_collector_is_executable(self) -> None:
-        current_evidence = self.baseline.split("## Evidence commands", 1)[1]
-        self.assertIn("scripts/ci/collect_live_merge_evidence.sh", current_evidence)
-        self.assertIn("scripts/ci/collect_live_merge_evidence.sh", self.agents)
-        self.assertNotEqual(0, EVIDENCE_SCRIPT.stat().st_mode & 0o111)
-
-    def test_evidence_collector_is_valid_bash(self) -> None:
-        result = subprocess.run(
-            ["bash", "-n", str(EVIDENCE_SCRIPT)],
-            cwd=ROOT,
-            check=False,
-            capture_output=True,
-            text=True,
-        )
-        self.assertEqual(0, result.returncode, result.stderr)
-
-    def test_current_change_requests_block_the_approval_verdict(self) -> None:
-        self.assertIn("as $current_change_requests", self.evidence)
-        self.assertIn("blocking_change_requests: $current_change_requests", self.evidence)
-        self.assertIn("($current_change_requests | length) == 0", self.evidence)
-
-    def test_every_unresolved_thread_remains_blocking_when_outdated(self) -> None:
-        self.assertIn("select(.isResolved == false)", self.evidence)
-        self.assertNotIn(
-            "select(.isResolved == false and .isOutdated == false)",
-            self.evidence,
-        )
-
-    def test_workflow_evidence_is_bound_to_pr_head_and_base(self) -> None:
-        for marker in (
-            ".number == ($pr[0].number)",
-            ".head.sha == $head",
-            ".base.sha == $base",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, self.evidence)
-        self.assertIn("workflow_runs_without_exact_pr_base_provenance", self.evidence)
-
-    def test_rules_are_evaluated_for_each_pull_requests_actual_base(self) -> None:
-        for marker in (
-            "BASE_REF=$(jq -r '.base.ref' \"$PR_JSON\")",
-            'BASE_REF_ENCODED=$(jq -rn --arg value "$BASE_REF" \'$value | @uri\')',
-            '"repos/$REPOSITORY/rules/branches/$BASE_REF_ENCODED?per_page=100"',
-            '--slurpfile rules "$EVIDENCE_DIR/pr-${PR}-branch-rules.json"',
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, self.evidence)
-        self.assertNotIn(
-            '--slurpfile rules "$EVIDENCE_DIR/main-branch-rules.json"',
-            self.evidence,
-        )
-
-    def test_verdict_materializes_only_after_inventory_and_final_state_stabilize(self) -> None:
-        for marker in (
-            "INVENTORY_DRAFT=$(jq -r --argjson pr \"$PR\"",
-            "PR_STATE=$(jq -r '.state' \"$PR_JSON\")",
-            "PR_DRAFT=$(jq -r '.draft' \"$PR_JSON\")",
-            "RECHECKED_STATE=$(jq -r '.state' \"$RECHECKED_PR_JSON\")",
-            "RECHECKED_DRAFT=$(jq -r '.draft' \"$RECHECKED_PR_JSON\")",
-            '"$PR_STATE" == "open"',
-            '"$RECHECKED_STATE" == "$PR_STATE"',
-            '"$RECHECKED_DRAFT" == "$PR_DRAFT"',
-            '"$PR_DRAFT" == "$INVENTORY_DRAFT"',
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, self.evidence)
-
-    def test_whole_generation_closes_against_fresh_pr_and_issue_inventories(self) -> None:
-        for marker in (
-            'open-pr-pages-rechecked.json',
-            'open-prs-rechecked.json',
-            'open-issue-pages-rechecked.json',
-            'open-issues-rechecked.json',
-            'INITIAL_PR_INVENTORY_PROJECTION=',
-            'FINAL_PR_INVENTORY_PROJECTION=',
-            'INITIAL_ISSUE_INVENTORY_PROJECTION=',
-            'FINAL_ISSUE_INVENTORY_PROJECTION=',
-            'Live inventory changed during evidence collection.',
-            'rm -f "$EVIDENCE_DIR"/pr-*-merge-verdict.json',
-            'evidence-generation.json',
-            'complete: true',
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, self.evidence)
-
-    def test_explicit_evidence_directory_must_start_empty(self) -> None:
-        for marker in (
-            'if [[ $# -gt 0 ]]; then',
-            'find "$EVIDENCE_DIR" -mindepth 1 -print -quit',
-            'Evidence directory must be empty:',
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, self.evidence)
-
-
-if __name__ == "__main__":
-    unittest.main()
+_legacy = load_historical_contract(
+    "legacy_live_gap_evidence_integrity_contract.py",
+    "_originweave_legacy_live_gap_evidence_integrity_contract",
+)
+export_public(_legacy, globals())
diff --git a/tests/test_product_completion_gap_contract.py b/tests/test_product_completion_gap_contract.py
index 970c9c189..e919ccbf5 100644
--- a/tests/test_product_completion_gap_contract.py
+++ b/tests/test_product_completion_gap_contract.py
@@ -1,790 +1,9 @@
-"""Regression contract for the dated commercial-completion gap baseline."""
+"""Discovery entry point for preserved commercial-completion history contracts."""
 
-from __future__ import annotations
+from _historical_baseline_contract_loader import export_public, load_historical_contract
 
-import pathlib
-import re
-import unittest
-from unittest.mock import patch
-
-from test_documentation_active_pr_evidence_contract import active_pr_row, bounded_section
-
-ROOT = pathlib.Path(__file__).resolve().parents[1]
-BASELINE = ROOT / "docs/product-technical-gap-baseline.md"
-
-
-class ProductCompletionGapContractTests(unittest.TestCase):
-    """Keep the exact repository snapshot and completion tracks reviewable."""
-
-    def test_ack_checkpoint_separates_local_repair_from_review_acceptance(self) -> None:
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "#### Published intent acknowledgment verification",
-            "#### Published repeated-control repair",
-        )
-        for marker in (
-            "46db0045904f0289738df843d0a2f179c26673d3", "34186280263",
-            "8eda96915dbbe4cc617f834267c7464689c2844d",
-            "1406/14894/19018/1552", "pre-consumption", "resolved",
-            "terminal SUCCESS", "CLEAN", "not shipped", "148 repository",
-            "10d5e1ff78d46bbad004d4e0971d0ffceac757fa", "superseded",
-            "6b6c90ed3919ea84b69527ab688a087eeb45224d",
-            "1105/11574/14787/1218", "34188785932", "server-sent",
-            "RED", "1010", "1011", "Actual Edge", "terminal SUCCESS",
-            "6b5241c164f5283f8dd51b1846ef0e4dacec0b29", "predecessor",
-            "0c077445", "594/4974/5980/754", "#293", "three command intents",
-            "3 September 2026", "viewport/DPR reset intent",
-            "current-head visual acceptance is unproven",
-            "Viewport", "DevicePixelRatio", "TimeZone", "ReducedMotion",
-        ):
-            self.assertIn(marker, current)
-
-    def test_repeated_control_checkpoint_keeps_operational_gaps_open(self) -> None:
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "#### Published repeated-control repair",
-            "#### Published deadline and Close-code repairs",
-        )
-        for marker in (
-            "3a8e6f4f89db2f53c144adb3351c153d89adca58", "34184829970",
-            "1104/11553/14758/1216", "20 closure", "64", "65th",
-            "not hosted acceptance", "process exit", "profile cleanup",
-            "not shipped", "fresh", "exhausted", "reused",
-        ):
-            self.assertIn(marker, current)
-
-    def test_close_code_checkpoint_does_not_claim_hosted_acceptance(self) -> None:
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "#### Published deadline and Close-code repairs",
-            "#### Transport-closure verification and test-integrity repair",
-        )
-        for marker in (
-            "8716b9d441960a112446c1f89ab417ad9abe28d2", "34183766437",
-            "1101/11492/14662/1218", "18 closure", "142 repository",
-            "not hosted acceptance", "repeated Ping/Pong", "1016–2999",
-            "process exit", "profile cleanup", "not shipped",
-        ):
-            self.assertIn(marker, current)
-
-    def test_closure_checkpoint_separates_verified_and_pending_heads(self) -> None:
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "### Previous verified cut: 2026-09-08",
-            "## Observed snapshot: 2026-08-29",
-        )
-        for marker in (
-            "d126242c7198c447d0fab7983d529441340fd1c9",
-            "07ef43ec71b6dbd8540629bf1df5a63b81541ee4",
-            "363a78e36e7690e9ed5bf49829567e00e2ec5d59",
-            "34179452950", "34180304951", "1090/11218/14328/1210",
-            "126 open pull requests", "114 Draft", "14 open non-PR issues",
-            "not protected-main acceptance", "operation-wide deadline",
-            "repeated Ping/Pong", "no package release",
-        ):
-            self.assertIn(marker, current)
-
-    def test_end_reply_checkpoint_binds_published_evidence(self) -> None:
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "#### Published session-end reply binding: 05:35 UTC",
-            "#### Published status-response repair: 04:45 UTC",
-        )
-        for owner, head, parent, coverage in (
-            (251, "924ad97551750d4a901ded38b89488cc5438e54f", "#250 `bbdc6ace`", "1057/10840/13873/1194"),
-            (252, "363a78e36e7690e9ed5bf49829567e00e2ec5d59", "#251 `924ad975`", "1064/10898/13947/1194"),
-        ):
-            row = active_pr_row(current, owner)
-            for marker in (head, parent, coverage, "Published; Draft"):
-                self.assertIn(marker, row)
-        for marker in (
-            "34085877650", "34087239755",
-            "9f2249637f31916acf9874baec724ddedfe621e5b14fdb6f4bd850620e89f307",
-            "not hosted acceptance", "retained receipts", "#255", "#292",
-        ):
-            self.assertIn(marker, current)
-
-    def test_historical_evidence_cannot_replace_end_reply_checkpoint(self) -> None:
-        text = BASELINE.read_text(encoding="utf-8")
-        current = bounded_section(
-            text, "#### Published session-end reply binding: 05:35 UTC",
-            "#### Published status-response repair: 04:45 UTC",
-        )
-        for owner in (251, 252):
-            row = active_pr_row(current, owner)
-            stale = text.replace(row, row.replace("Published; Draft", "Local only"), 1)
-            with patch.object(pathlib.Path, "read_text", return_value=stale + row):
-                with self.assertRaises(AssertionError):
-                    self.test_end_reply_checkpoint_binds_published_evidence()
-
-    def test_status_repair_checkpoint_binds_published_evidence(self) -> None:
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "#### Published status-response repair: 04:45 UTC",
-            "#### Published semantic-action adoption: 04:11 UTC",
-        )
-        for marker in (
-            "bbdc6ace7a5932adf24836700f806850e6b230bc",
-            "Published; Draft",
-            "1043/10711/13717/1188",
-            "34084134654",
-            "e1fccefc6b56eabe653ae41377fbcec0aa89b4be1ada92dba73d3ef87e841029",
-            "not hosted acceptance",
-        ):
-            self.assertIn(marker, current)
-
-    def test_historical_evidence_cannot_replace_status_repair(self) -> None:
-        text = BASELINE.read_text(encoding="utf-8")
-        current = bounded_section(
-            text, "#### Published status-response repair: 04:45 UTC",
-            "#### Published semantic-action adoption: 04:11 UTC",
-        )
-        stale = text.replace(current, current.replace("Published; Draft", "Local only"), 1)
-        with patch.object(pathlib.Path, "read_text", return_value=stale + current):
-            with self.assertRaises(AssertionError):
-                self.test_status_repair_checkpoint_binds_published_evidence()
-
-    def test_current_snapshot_checks_do_not_route_by_phrase_substrings(self) -> None:
-        """Current-snapshot assertions must not depend on count-word substrings."""
-        source = pathlib.Path(__file__).read_text(encoding="utf-8")
-        brittle_condition = '"pull requests" in phrase or ' + '"draft" in phrase'
-        self.assertNotIn(brittle_condition, source)
-
-    def test_baseline_records_current_inventory_and_completion_issues(self) -> None:
-        """The dated baseline must not retain superseded queue counts or omit buyer tracks."""
-        text = BASELINE.read_text(encoding="utf-8")
-        end_marker = "#### 2026-08-29 maintenance-loop record"
-        self.assertIn(end_marker, text)
-        current = text.split("## Observed snapshot: ", 1)[1].split(end_marker, 1)[0]
-
-        for phrase in (
-            "108 open pull requests",
-            "24 non-draft",
-            "84 draft",
-            "2026-08-28 116-PR snapshot",
-        ):
-            with self.subTest(phrase=phrase):
-                self.assertIn(phrase, current)
-
-        for phrase in (
-            "#198",
-            "#199",
-            "#200",
-            "#201",
-            "#202",
-            "#203",
-            "Shrink the open-PR queue",
-            "durable WARC/PROV replay",
-            "stable BAP/MCP runtime API",
-            "signed cross-platform Chromium distribution",
-            "enterprise control and experience plane",
-            "commercial acceptance gate",
-        ):
-            with self.subTest(phrase=phrase):
-                self.assertIn(phrase, text)
-
-        self.assertNotIn("Shrink the 109-PR queue", text)
-        self.assertNotIn(
-            "current-head hosted CI, security, Noema, scheduler, and OpenCode workflows were still regenerating at the recheck",
-            text,
-        )
-
-        for stale_phrase in (
-            "100 open pull requests",
-            "22 non-draft",
-            "78 draft",
-            "148 open pull requests",
-            "79 draft PRs",
-            "40 non-draft",
-            "110 draft",
-            "150 open pull requests",
-            "prior 150-PR snapshot",
-            "128 open pull requests",
-            "74 draft",
-            "115 open pull requests",
-            "31 non-draft",
-        ):
-            with self.subTest(stale_phrase=stale_phrase):
-                self.assertNotIn(stale_phrase, current)
-
-    def test_active_github_approval_rule_is_not_documented_as_bypassable(self) -> None:
-        """An active counted-approval rule must stop merge without an eligible approver."""
-        text = BASELINE.read_text(encoding="utf-8")
-
-        self.assertIn("eligible non-author", text)
-        self.assertIn("reviewer-provisioning gap", text)
-        self.assertNotIn("owner-directed administrative merge", text)
-
-    def test_current_snapshot_records_the_pr_284_admin_bypass_incident(self) -> None:
-        """A bypassed main update must not be presented as a policy-compliant merge."""
-        text = BASELINE.read_text(encoding="utf-8")
-        current = text.split("## Current live delivery state", 1)[1].split(
-            "## Observed snapshot: ", 1
-        )[0]
-
-        for marker in (
-            "87c4daa1830bac5a5228b6036752ad5633232085",
-            "#284",
-            "61bcf88c960c6c437ccd29b3fbb73cd4325f9e5a",
-            "rule-suite `3948421709`",
-            "`result: bypass`",
-            "#215",
-            "fresh native Rust and production-coverage checks now succeed",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, current)
-
-        self.assertNotIn("through #280", current)
-
-    def test_current_snapshot_records_repaired_webdriver_bidi_lineage(self) -> None:
-        """Only the latest cut may establish current foundation and transport heads."""
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "#### Connection-bound text responses: 14:54 UTC",
-            "#### Session repair and child adoption: 13:35 UTC",
-        )
-        self.assertIn("#195 `63997bcf555e2c5c8e91ba287734ffba3837a1b7`", current)
-        self.assertIn("#267 `3346d8ecc72932b98ec495d9cc52d6e5727c3064`", current)
-        self.assertIn("#268 `e567af9e678fd4791776df795e89ed666975e6c2`", current)
-        for marker in (
-            "4632f2df", "d6889c80", "e1188c86",
-            "34040202356", "34040306372", "queued",
-            "pointer and status", "not product-browser acceptance",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, current)
-
-    def test_latest_executable_queue_uses_current_ready_roots(self) -> None:
-        text = BASELINE.read_text(encoding="utf-8")
-        current = bounded_section(
-            text, "#### Published session-end reply binding: 05:35 UTC",
-            "#### Published status-response repair: 04:45 UTC",
-        )
-        roots = [line for line in current.splitlines() if line.startswith("Ready roots:")]
-        self.assertEqual(len(roots), 1)
-        self.assertEqual(
-            set(re.findall(r"#(\d+)", roots[0])),
-            {"37", "50", "166", "219", "220", "229", "238", "240", "272", "274", "285", "287"},
-        )
-        self.assertIn("## Historical next executable queue", text)
-        self.assertNotIn("## Next executable queue", text)
-
-    def test_pointer_checkpoint_records_published_receipt_repair(self) -> None:
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "##### Pointer receipt follow-up: 15:28 UTC",
-            "#### Session repair and child adoption: 13:35 UTC",
-        )
-        for marker in (
-            "#257 `9451fd8a23dec95b31749376bc78c2eaca977fe8`",
-            "#258 `5417ce32ed957aa166807f1023647caccc2920cb`",
-            "8193fcd5", "d9396f05", "588fe731", "0234b587",
-            "2 → 2 → 1", "34041977863", "34042223733", "queued",
-            "not product-browser acceptance", "outbound session authority",
-            "#249", "#250", "descendant adoption",
-        ):
-            self.assertIn(marker, current)
-
-    def test_historical_text_cannot_supply_missing_pointer_receipt_head(self) -> None:
-        text = BASELINE.read_text(encoding="utf-8")
-        marker = "#258 `5417ce32ed957aa166807f1023647caccc2920cb`"
-        mutated = text.replace(marker, "response head removed", 1)
-        self.assertNotEqual(mutated, text)
-        mutated += "\nHistorical evidence: " + marker
-        with patch.object(pathlib.Path, "read_text", return_value=mutated):
-            with self.assertRaises(AssertionError):
-                self.test_pointer_checkpoint_records_published_receipt_repair()
-
-    def test_descendant_checkpoint_records_exact_receipt_adoptions(self) -> None:
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "##### Receipt descendant adoption: 16:17 UTC",
-            "#### Session repair and child adoption: 13:35 UTC",
-        )
-        for marker in (
-            "180c168ecbdcd5eb7a4ad14ab4a53e8670646bf7",
-            "3807aabeb22f9622610c3c8d504d1c686d25d896",
-            "ba100fbc39e1ac4f10ee4faade38418551bb8298",
-            "121578a43adda12221a9ca9ab8ada0a4fd03efef",
-            "34043222194", "34043664230", "34044193429", "34044857929",
-            "queued", "144 Python", "subscription response provenance",
-            "protected-main asset", "not product-browser acceptance",
-        ):
-            self.assertIn(marker, current)
-
-    def test_historical_text_cannot_supply_missing_descendant_head(self) -> None:
-        text = BASELINE.read_text(encoding="utf-8")
-        marker = "121578a43adda12221a9ca9ab8ada0a4fd03efef"
-        mutated = text.replace(marker, "subscription head removed", 1)
-        self.assertNotEqual(mutated, text)
-        mutated += "\nHistorical evidence: " + marker
-        with patch.object(pathlib.Path, "read_text", return_value=mutated):
-            with self.assertRaises(AssertionError):
-                self.test_descendant_checkpoint_records_exact_receipt_adoptions()
-
-    def test_subscription_checkpoint_records_verified_repair_and_visual_gap(self) -> None:
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "##### Subscription response repair: 16:37 UTC",
-            "#### Session repair and child adoption: 13:35 UTC",
-        )
-        for marker in (
-            "46ae62aa31e35c702cd61c16322d05c7a9c35da1", "122ca139", "8b1508c8",
-            "0/2", "14 focused", "144 Python", "1222/12824/16453/1418",
-            "34045953423", "queued", "Mac is locked", "visual inspection remains pending",
-            "outbound session binding", "protected-main asset", "not product-browser acceptance",
-        ):
-            self.assertIn(marker, current)
-
-    def test_observation_safeguards_bind_each_published_owner(self) -> None:
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "#### Published observation safeguards: 03:00 UTC",
-            "#### Published response and observation: 01:45 UTC",
-        )
-        for owner, head, parent, coverage in (
-            (270, "8eda96915dbbe4cc617f834267c7464689c2844d", "3df2a631", "1346/14113/17928/1464"),
-            (271, "b0410ae92bd20eaf31d09b7d49390e13cb045999", "8eda9691", "1393/14770/18899/1546"),
-        ):
-            row = active_pr_row(current, owner)
-            for marker in ("Published; Draft", parent, coverage, f"/commit/{head}"):
-                self.assertIn(marker, row)
-        for marker in (
-            "29cd0d66", "14efb678", "148 Python", "34076117534", "34077987302",
-            "queued", "actual visual inspection", "not product-browser acceptance",
-            "original connection", "status receipts", "#195/#279",
-        ):
-            self.assertIn(marker, current)
-
-    def test_action_adoption_checkpoint_binds_published_owners(self) -> None:
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "#### Published semantic-action adoption: 04:11 UTC",
-            "#### Published observation safeguards: 03:00 UTC",
-        )
-        for owner, head, parent, coverage in (
-            (93, "82056d13aa94c106060b84ee76be56fcb7787fc8", "b0410ae9", "1403/14852/18976/1552"),
-            (95, "6b29d890245ed2f612c2998198f4e8c8a06da312", "82056d13", "1408/14897/19022/1552"),
-            (96, "cbabf55c6a25b979fa0d9e3c1677338665975ab7", "6b29d890", "1410/14908/19030/1552"),
-        ):
-            row = active_pr_row(current, owner)
-            for marker in ("Published; Draft", parent, coverage, f"/commit/{head}"):
-                self.assertIn(marker, row)
-        for marker in (
-            "34079739018", "34080772063", "34081979602", "queued",
-            "actual visual inspection", "not product-browser acceptance",
-            "original connection", "status receipts", "publish = false",
-        ):
-            self.assertIn(marker, current)
-
-    def test_historical_rows_cannot_replace_action_adoption(self) -> None:
-        text = BASELINE.read_text(encoding="utf-8")
-        current = bounded_section(text, "#### Published semantic-action adoption: 04:11 UTC", "#### Published observation safeguards: 03:00 UTC")
-        for owner in (93, 95, 96):
-            row = active_pr_row(current, owner)
-            for replacement in ("", row.replace("Published; Draft", "Local only; Draft")):
-                mutated = text.replace(row, replacement, 1) + "\nHistorical evidence:\n" + row
-                with patch.object(pathlib.Path, "read_text", return_value=mutated):
-                    with self.assertRaises(AssertionError):
-                        self.test_action_adoption_checkpoint_binds_published_owners()
-
-    def test_historical_rows_cannot_replace_observation_safeguards(self) -> None:
-        text = BASELINE.read_text(encoding="utf-8")
-        current = bounded_section(text, "#### Published observation safeguards: 03:00 UTC", "#### Published response and observation: 01:45 UTC")
-        for owner in (270, 271):
-            row = active_pr_row(current, owner)
-            for replacement in ("", row.replace("Published; Draft", "Local only; Draft")):
-                mutated = text.replace(row, replacement, 1) + "\nHistorical evidence:\n" + row
-                with patch.object(pathlib.Path, "read_text", return_value=mutated):
-                    with self.assertRaises(AssertionError):
-                        self.test_observation_safeguards_bind_each_published_owner()
-
-    def test_response_observation_checkpoint_binds_each_published_owner(self) -> None:
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "#### Published response and observation: 01:45 UTC",
-            "#### Published input descendants: 00:40 UTC",
-        )
-        for owner, head, parent, coverage in (
-            (268, "ff27220cb5eb4d11ca1dc5614a4181e1a397a3f1", "ebd507ae", "1325/13907/17682/1456"),
-            (269, "3df2a631bacd7109b3982fdd7ac599d0bd92a589", "ff27220c", "1338/14025/17843/1460"),
-        ):
-            row = active_pr_row(current, owner)
-            for marker in ("Published; Draft", parent, coverage):
-                self.assertIn(marker, row)
-            self.assertIn(f"[`{head[:8]}`](https://github.com/ContextualWisdomLab/OriginWeave/commit/{head})", row)
-        for marker in ("658fb676", "49d18f5f", "147 Python", "34072645796", "34073733364", "34073733355", "queued", "actual visual inspection", "not product-browser acceptance", "#270"):
-            self.assertIn(marker, current)
-
-    def test_historical_rows_cannot_replace_response_observation_checkpoint(self) -> None:
-        text = BASELINE.read_text(encoding="utf-8")
-        current = bounded_section(text, "#### Published response and observation: 01:45 UTC", "#### Published input descendants: 00:40 UTC")
-        for owner in (268, 269):
-            row = active_pr_row(current, owner)
-            for replacement in ("", row.replace("Published; Draft", "Local only; Draft")):
-                mutated = text.replace(row, replacement, 1) + "\nHistorical evidence:\n" + row
-                with patch.object(pathlib.Path, "read_text", return_value=mutated):
-                    with self.assertRaises(AssertionError):
-                        self.test_response_observation_checkpoint_binds_each_published_owner()
-
-    def test_input_descendants_bind_publication_and_coverage_to_each_owner(self) -> None:
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "#### Published input descendants: 00:40 UTC",
-            "### Latest verified cut: 2026-09-06",
-        )
-        for owner, head, parent, coverage in (
-            (265, "e94a2372fe3771f9ddf70291d34fc9a7e4770ec9", "43395711", "1297/13618/17315/1442"),
-            (266, "e3885f69df2cf3899184209efdee5b11bba1bd86", "e94a2372", "1310/13766/17524/1452"),
-            (267, "ebd507ae56c3064e3cae5566502f539c20618a8f", "e3885f69", "1318/13852/17610/1456"),
-        ):
-            row = active_pr_row(current, owner)
-            for marker in ("Published; Draft", parent, coverage):
-                self.assertIn(marker, row)
-            self.assertIn(
-                f"[`{head[:8]}`](https://github.com/ContextualWisdomLab/OriginWeave/commit/{head})",
-                row,
-            )
-        for marker in (
-            "e7fb1527", "009f9a41", "4e020e16", "23 focused", "145 Python",
-            "34068277243", "34068882527", "34070121583", "queued",
-            "zero unresolved review threads", "one counted approval", "seven required workflows",
-            "actual screenshots", "not product-browser acceptance", "own checks and visual inspection",
-            "#268", "protected-main asset preservation", "status receipt",
-        ):
-            self.assertIn(marker, current)
-
-    def test_historical_rows_cannot_hide_current_input_evidence_changes(self) -> None:
-        text = BASELINE.read_text(encoding="utf-8")
-        current = bounded_section(
-            text, "#### Published input descendants: 00:40 UTC",
-            "### Latest verified cut: 2026-09-06",
-        )
-        for owner in (265, 266, 267):
-            row = active_pr_row(current, owner)
-            for replacement in ("", row.replace("Published; Draft", "Local only; Draft")):
-                with self.subTest(owner=owner, replacement=replacement):
-                    mutated = text.replace(row, replacement, 1) + "\nHistorical evidence:\n" + row
-                    self.assertNotEqual(text, mutated)
-                    with patch.object(pathlib.Path, "read_text", return_value=mutated):
-                        with self.assertRaises(AssertionError):
-                            self.test_input_descendants_bind_publication_and_coverage_to_each_owner()
-
-    def test_published_descendants_bind_evidence_to_current_owner_rows(self) -> None:
-        current = bounded_section(
-            BASELINE.read_text(encoding="utf-8"),
-            "#### Published subscription descendants: 23:27 UTC",
-            "#### Connection-bound text responses: 14:54 UTC",
-        )
-        for owner, head, parent, coverage in (
-            (263, "4868d3e9f19133ac3382ee8532878aef27468893", "46ae62aa", "1244/13053/16729/1422"),
-            (264, "433957117ad9e29b26715b062f5adcc9789744ba", "4868d3e9", "1282/13461/17140/1440"),
-        ):
-            row = active_pr_row(current, owner)
-            for marker in ("Published; Draft", head, parent, coverage):
-                self.assertIn(marker, row)
-            self.assertIn(
-                f"[`{head[:8]}`](https://github.com/ContextualWisdomLab/OriginWeave/commit/{head})",
-                row,
-            )
-        for marker in (
-            "zero unresolved review threads", "supersedes the older sole-#147",
-            "12 Ready candidates remain BLOCKED", "one counted approval", "seven required workflows",
-            "13 focused", "145 Python", "actual in-app screenshots", "not product-browser acceptance",
-            "34065055213", "34066516991", "34066516992", "queued",
-            ".github#1929", "protected-main asset preservation", "own checks and visual inspection",
-        ):
-            self.assertIn(marker, current)
-
-    def test_historical_rows_cannot_hide_current_descendant_evidence_changes(self) -> None:
-        text = BASELINE.read_text(encoding="utf-8")
-        current = bounded_section(
-            text,
-            "#### Published subscription descendants: 23:27 UTC",
-            "#### Connection-bound text responses: 14:54 UTC",
-        )
-        for owner in (263, 264):
-            row = active_pr_row(current, owner)
-            for replacement in ("", row.replace("Published; Draft", "Local only; Draft")):
-                with self.subTest(owner=owner, replacement=replacement):
-                    mutated = text.replace(row, replacement, 1) + "\nHistorical evidence:\n" + row
-                    self.assertNotEqual(text, mutated)
-                    with patch.object(pathlib.Path, "read_text", return_value=mutated):
-                        with self.assertRaises(AssertionError):
-                            self.test_published_descendants_bind_evidence_to_current_owner_rows()
-
-    def test_historical_text_cannot_supply_missing_subscription_repair_head(self) -> None:
-        text = BASELINE.read_text(encoding="utf-8")
-        marker = "46ae62aa31e35c702cd61c16322d05c7a9c35da1"
-        mutated = text.replace(marker, "subscription repair head removed", 1)
-        self.assertNotEqual(mutated, text)
-        mutated += "\nHistorical evidence: " + marker
-        with patch.object(pathlib.Path, "read_text", return_value=mutated):
-            with self.assertRaises(AssertionError):
-                self.test_subscription_checkpoint_records_verified_repair_and_visual_gap()
-
-    def test_historical_prose_cannot_hide_latest_root_or_lineage_removal(self) -> None:
-        text = BASELINE.read_text(encoding="utf-8")
-        cases = (
-            (
-                "#195 `63997bcf555e2c5c8e91ba287734ffba3837a1b7`",
-                "foundation evidence removed",
-                self.test_current_snapshot_records_repaired_webdriver_bidi_lineage,
-            ),
-            (
-                "Ready roots: #37, #50,",
-                "Ready roots: #37,",
-                self.test_latest_executable_queue_uses_current_ready_roots,
-            ),
-        )
-        for original, replacement, check in cases:
-            with self.subTest(original=original):
-                mutated = text.replace(original, replacement, 1)
-                self.assertNotEqual(mutated, text)
-                with patch.object(pathlib.Path, "read_text", return_value=mutated):
-                    with self.assertRaises(AssertionError):
-                        check()
-
-    def test_historical_snapshot_preserves_webdriver_bidi_lineage(self) -> None:
-        """The active stack must retain exact heads and the macOS race boundary."""
-        text = BASELINE.read_text(encoding="utf-8")
-        current = text.split("## Current live delivery state", 1)[1].split(
-            "## Observed snapshot: ", 1
-        )[0]
-
-        for marker in (
-            "#195 is Draft at exact head `48eb2d23009c1c804520dd5efcd0d4d072aacef1`",
-            "50 consecutive focused regression passes",
-            "#242 is Draft at exact head `2d0e9f69df9ade21d8e8e3d807c3ff644d83b310`",
-            "PR #243 adopts that corrected parent by ordinary merge at `97fab641ed9d76e6c515eadcef0629edfc8064a3`",
-            "#93 is Draft at exact head `0664f0452cb329cd692cce7f61f9001652abfda2`",
-            "SemanticNodeActionBinding",
-            "does not authorize policy or execute input",
-            "#95 is Draft at exact head `97aa0f2e340ee6fd920d0418f97af276b190554f`",
-            "Only `Decision::Allow`",
-            "`NotAdmitted`",
-            "#96 is Draft at exact head `b7ba8dd1433410cee43084a73e31816da841b2a2`",
-            "callback is never invoked",
-            "#101 is Draft at exact head `bc810b121bb0303f55afa8777a23cc0f9748c1db`",
-            "known-disabled interactive action",
-            "#102 is Draft at exact head `a123c55d4839dae1db7e6671f7d4d158c7cfd9db`",
-            "ObservationAuthorityMismatch",
-            "#103 is Draft at exact head `8b3416169346fa04b53c915b813d55ccf47d1876`",
-            "browser authority first",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, current)
-
-        self.assertIn(
-            "| #195 | Draft | `6922dd98779e8f8aad132a3b1f563d7ba6e6d070` | "
-            "`48eb2d23009c1c804520dd5efcd0d4d072aacef1` |",
-            text,
-        )
-        self.assertIn(
-            "| #93 | Draft | `802ec806cdd4560eab48c484f435766ecabda353` | "
-            "`0664f0452cb329cd692cce7f61f9001652abfda2` |",
-            text,
-        )
-        self.assertIn(
-            "| #242 | Draft | `48eb2d23009c1c804520dd5efcd0d4d072aacef1` | "
-            "`55fef0c3fae1724eddada53e52c4a0311f509aa3` |",
-            text,
-        )
-        self.assertIn(
-            "| #95 | Draft | `0664f0452cb329cd692cce7f61f9001652abfda2` | "
-            "`97aa0f2e340ee6fd920d0418f97af276b190554f` |",
-            text,
-        )
-        self.assertIn(
-            "| #96 | Draft | `97aa0f2e340ee6fd920d0418f97af276b190554f` | "
-            "`b7ba8dd1433410cee43084a73e31816da841b2a2` |",
-            text,
-        )
-        self.assertIn(
-            "| #101 | Draft | `b7ba8dd1433410cee43084a73e31816da841b2a2` | "
-            "`bc810b121bb0303f55afa8777a23cc0f9748c1db` |",
-            text,
-        )
-        self.assertIn(
-            "| #102 | Draft | `bc810b121bb0303f55afa8777a23cc0f9748c1db` | "
-            "`a123c55d4839dae1db7e6671f7d4d158c7cfd9db` |",
-            text,
-        )
-        self.assertIn(
-            "| #103 | Draft | `a123c55d4839dae1db7e6671f7d4d158c7cfd9db` | "
-            "`8b3416169346fa04b53c915b813d55ccf47d1876` |",
-            text,
-        )
-
-    def test_current_snapshot_records_origin_bound_socket_ports(self) -> None:
-        """The live network root must retain its exact port-authority repair."""
-        text = BASELINE.read_text(encoding="utf-8")
-        current = text.split("## Current live delivery state", 1)[1].split(
-            "## Observed snapshot: ", 1
-        )[0]
-
-        for marker in (
-            "PR #50 is Ready at exact head `ad87cfea59db711cb29ef90559790ba77e22029f`",
-            "Fresh independent verification passed all seven fresh-resolution tests and 152 Python contracts",
-            "origin-approved IP could be paired with a different service port",
-            "binds the socket port to the effective scheme-host-port origin",
-            "function/line/region/branch coverage pass locally at 100%",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, current)
-
-    def test_current_snapshot_records_content_length_persistence_repair(self) -> None:
-        """A declared body boundary must not be confused with transport closure."""
-        text = BASELINE.read_text(encoding="utf-8")
-        current = text.split("## Current live delivery state", 1)[1].split(
-            "## Observed snapshot: ", 1
-        )[0]
-
-        for marker in (
-            "PR #37 is Ready at exact head `1e2f41072854edcdbaf0f9ecf14697a3bfd62195`",
-            "returns after the exact declared bytes",
-            "already-buffered surplus remains fail-closed",
-            "does not require TLS EOF",
-        ):
-            with self.subTest(marker=marker):
-                self.assertIn(marker, current)
-
-    def test_changelog_marks_superseded_warc_head_as_historical(self) -> None:
-        """A predecessor WARC head must not look like the current exact evidence."""
-        text = (ROOT / "CHANGELOG.md").read_text(encoding="utf-8")
-
-        self.assertIn("`d83748a70bd1b16dbfec46007fe02989ba6ce188` was superseded", text)
-        self.assertIn("0341079331f9cea669eb9a5cc21842fd6027431e", text)
-
-    def test_baseline_records_central_required_workflow_failures(self) -> None:
-        """The baseline must preserve exact central workflow provenance and failures."""
-        text = BASELINE.read_text(encoding="utf-8")
-
-        for phrase in (
-            "central `.github` repository",
-            "repository ID `1274066402`",
-            "close-empty-pr",
-            "opencode-review",
-            "pr-review-merge-scheduler",
-            "security-scan",
-            "strix",
-            "sast-semgrep",
-            "noema-review",
-            "OpenCode current-head verdict",
-            "Strix provider/backend",
-            "internal server error",
-            "33177641855",
-            "33177641888",
-            "33182772296",
-            "MODEL_OUTPUT_UNAVAILABLE",
-            "model pool exhausted",
-            "33182749298",
-            "#1391",
-            "e4ba6b599cd1e50d0139762885682607b731655d",
-            "did not prove missing workflow identities",
-            "does not authorize bypass",
-        ):
-            with self.subTest(phrase=phrase):
-                self.assertIn(phrase, text)
-
-    def test_same_day_prior_inventory_is_bound_to_the_maintenance_record(self) -> None:
-        """A same-day comparison must retain its exact prior observation in the record."""
-        text = BASELINE.read_text(encoding="utf-8")
-        record_end = "#### Current exact-head active PR evidence"
-        self.assertIn(record_end, text)
-        record = text.split("#### 2026-08-29 maintenance-loop record", 1)[1].split(
-            record_end, 1
-        )[0]
-        self.assertIn("115 open pull requests (31 ready, 84 draft)", record)
-
-    def test_historical_pr_219_head_is_a_full_exact_sha(self) -> None:
-        """The retained PR #219 regression anchor must identify its real commit."""
-        text = BASELINE.read_text(encoding="utf-8")
-
-        self.assertIn(
-            "| #219 | Ready | `0841d2ab3d8b5e60a03c0a8e818cf438e2716829` | "
-            "`8145d40f1b028a8f4dc7e7da47ac89bb9e5bb2c7` |",
-            text,
-        )
-
-    def test_issue_table_distinguishes_open_issues_from_governance_signals(self) -> None:
-        """The table total must distinguish product issues from governance signals."""
-        text = BASELINE.read_text(encoding="utf-8")
-        table_end = "## Buyer-visible and technical gap matrix"
-        self.assertIn(table_end, text)
-        table = text.split("### Open issues and governance signals", 1)[1].split(
-            table_end, 1
-        )[0]
-        self.assertIn("11 open issues plus 2 governance signals", table)
-        self.assertIn("Issue or signal", table)
-
-    def test_evidence_commands_reproduce_inventory_checks_and_review_state(self) -> None:
-        """The baseline links to the executable procedure without duplicating it."""
-        text = BASELINE.read_text(encoding="utf-8")
-        evidence = text.split("## Evidence commands", 1)[1].split("\n## ", 1)[0]
-        script = (ROOT / "scripts" / "ci" / "collect_live_merge_evidence.sh").read_text(
-            encoding="utf-8"
-        )
-
-        self.assertIn("[collect_live_merge_evidence.sh]", evidence)
-        self.assertNotIn("```bash", evidence)
-
-        for phrase in (
-            '"repos/$REPOSITORY/pulls?state=open&per_page=100"',
-            "set -euo pipefail",
-            'EVIDENCE_DIR="$(mktemp -d /tmp/originweave-evidence.XXXXXX)"',
-            '"$EVIDENCE_DIR/open-pr-pages.json"',
-            "jq '[.[][]]' \"$EVIDENCE_DIR/open-pr-pages.json\"",
-            '"repos/$REPOSITORY/issues?state=open&per_page=100"',
-            '"$EVIDENCE_DIR/open-issue-pages.json"',
-            'map(select(has("pull_request") | not))',
-            "open_non_pr_issues",
-            '"repos/$REPOSITORY/pulls/$PR"',
-            '"repos/$REPOSITORY/commits/$HEAD_SHA/check-runs?per_page=100"',
-            '"repos/$REPOSITORY/commits/$HEAD_SHA/statuses?per_page=100"',
-            '"repos/$REPOSITORY/pulls/$PR/reviews?per_page=100"',
-            '"repos/$REPOSITORY/actions/runs?head_sha=$HEAD_SHA&per_page=100"',
-            "check_runs: [$checks[][].check_runs[]?],",
-            "legacy_statuses: [$statuses[][][]?]",
-            "workflow_runs: $exact_pr_base_workflow_runs,",
-            "reviewThreads(first: 100, after: $endCursor)",
-            "rules/branches/main?per_page=100",
-            '"$EVIDENCE_DIR/main-branch-rule-pages.json"',
-            '"$EVIDENCE_DIR/collaborator-pages.json"',
-            '"$EVIDENCE_DIR/collaborators.json"',
-            '"$EVIDENCE_DIR/pr-${PR}-merge-verdict.json.tmp"',
-            '.state == "APPROVED"',
-            ".submitted_at != null",
-            ".commit_id == $head",
-            "group_by(.reviewer)",
-            "required_approving_review_count",
-            "require_last_push_approval",
-            "last_push_approval_authority",
-            '"github_rule_evaluation_required"',
-            "if $require_last_push_approval == true then false",
-            "$pr[0].user.login",
-            '.type == "workflows"',
-            ".parameters.workflows",
-            "required_status_checks",
-            '"$EVIDENCE_DIR/pr-${PR}-merge-verdict.json"',
-            "for ATTEMPT in 1 2 3; do",
-            "RECHECKED_HEAD_SHA=",
-            "RECHECKED_BASE_SHA=",
-            '"$RECHECKED_BASE_SHA" == "$BASE_SHA" &&',
-        ):
-            with self.subTest(phrase=phrase):
-                self.assertIn(phrase, script)
-
-        self.assertNotIn("while :; do", script)
-        self.assertNotIn("/tmp/originweave-open-pr", script)
-        self.assertNotIn("check_runs: [$checks[]?.check_runs[]?],", script)
-        self.assertNotIn("legacy_statuses: [$statuses[][]?]", script)
-        self.assertNotIn("workflow_runs: [$workflow_runs[]?.workflow_runs[]?],", script)
-        self.assertNotIn("$reviews[][]?\n          | select(.state", script)
-        self.assertNotIn("head-commit.json", script)
-        self.assertNotIn("$head_commit[0].committer.login", script)
-        self.assertNotIn("$head_commit[0].author.login", script)
-
-
-if __name__ == "__main__":
-    unittest.main()
+_legacy = load_historical_contract(
+    "legacy_product_completion_gap_contract.py",
+    "_originweave_legacy_product_completion_gap_contract",
+)
+export_public(_legacy, globals())
diff --git a/tests/test_product_gap_baseline_navigation_contract.py b/tests/test_product_gap_baseline_navigation_contract.py
new file mode 100644
index 000000000..8f9e804ba
--- /dev/null
+++ b/tests/test_product_gap_baseline_navigation_contract.py
@@ -0,0 +1,77 @@
+"""Buyer-facing navigation contract for the live product-gap decision surface."""
+
+from __future__ import annotations
+
+from pathlib import Path
+import unittest
+
+ROOT = Path(__file__).resolve().parents[1]
+BASELINE = ROOT / "docs/product-technical-gap-baseline.md"
+ARCHIVE = ROOT / "docs/evidence/product-technical-gap-baseline-through-2026-09-09.md"
+ARCHIVE_CHANGELOG = ROOT / "docs/evidence/CHANGELOG-through-2026-09-09.md"
+CHANGELOG = ROOT / "CHANGELOG.md"
+LOADER = ROOT / "tests/_historical_baseline_contract_loader.py"
+
+
+class ProductGapBaselineNavigationContractTests(unittest.TestCase):
+    """Keep current decisions reachable without discarding immutable history."""
+
+    def test_current_baseline_is_bounded_and_reaches_buyer_matrix_early(self) -> None:
+        text = BASELINE.read_text(encoding="utf-8")
+        lines = text.splitlines()
+
+        self.assertLessEqual(len(lines), 220)
+        self.assertLessEqual(len(text.encode("utf-8")), 24_000)
+        self.assertIn("## Observed delivery cut — 2026-09-09 12:57 UTC", text)
+        self.assertIn("## Buyer gap matrix", text)
+        self.assertIn("## Evidence and history index", text)
+        self.assertLess(lines.index("## Buyer gap matrix"), 90)
+        self.assertNotIn("### Previous verified cut:", text)
+        self.assertIn(ARCHIVE.name, text)
+
+    def test_observation_receipt_is_self_inclusive_and_not_continuously_live(self) -> None:
+        text = BASELINE.read_text(encoding="utf-8")
+
+        self.assertNotIn("## Current exact observation", text)
+        self.assertIn("This is a dated observation receipt, not a continuously live counter.", text)
+        self.assertIn("130 open pull requests: 12 Ready/non-draft and 118 Draft", text)
+        self.assertIn("includes this #309 Draft successor", text)
+        self.assertIn("Live GitHub state supersedes this cut after 2026-09-09 12:57 UTC.", text)
+
+    def test_live_changelog_uses_the_same_dated_inventory_receipt(self) -> None:
+        changelog = CHANGELOG.read_text(encoding="utf-8")
+        inventory = [
+            line
+            for line in changelog.splitlines()
+            if line.startswith("- Current delivery inventory:")
+        ]
+
+        self.assertEqual(1, len(inventory))
+        self.assertIn("130 open pull requests (12 ready, 118 draft)", inventory[0])
+        self.assertIn("14 open non-PR issues", inventory[0])
+        self.assertIn("Observed 2026-09-09 12:57 UTC", inventory[0])
+        self.assertNotIn("131 open pull requests (14 ready, 117 draft)", inventory[0])
+
+    def test_historical_dossier_is_preserved_outside_the_decision_surface(self) -> None:
+        self.assertTrue(ARCHIVE.is_file())
+        archive = ARCHIVE.read_text(encoding="utf-8")
+
+        self.assertGreater(len(archive.encode("utf-8")), 180_000)
+        self.assertIn("### Previous verified cut: 2026-09-08", archive)
+        self.assertIn("## Observed snapshot: 2026-08-29", archive)
+        self.assertIn("#### Published session-end reply binding: 05:35 UTC", archive)
+
+    def test_legacy_changelog_input_is_immutable_and_separate_from_live_changelog(self) -> None:
+        self.assertTrue(ARCHIVE_CHANGELOG.is_file())
+        archived = ARCHIVE_CHANGELOG.read_text(encoding="utf-8")
+        live = CHANGELOG.read_text(encoding="utf-8")
+        loader = LOADER.read_text(encoding="utf-8")
+
+        self.assertIn("131 open pull requests (14 ready, 117 draft)", archived)
+        self.assertIn("130 open pull requests (12 ready, 118 draft)", live)
+        self.assertIn("ARCHIVE_CHANGELOG", loader)
+        self.assertIn("module.CHANGELOG = ARCHIVE_CHANGELOG", loader)
+
+
+if __name__ == "__main__":
+    unittest.main()