docs: establish canonical acquisition documentation on current main

[seonghobae]

Purpose

Establish the single canonical DiskSage product/acquisition documentation graph without importing unrelated product, coverage, release, or obsolete repair-workflow changes.

Current exact state

• Exact source head: 4db841159b2f6b8c30bfc53d5aeb0365261d19c0.
• PR base snapshot: 4b2f5d30c1f1961e017a84aa1c785d3f1b1bb9af.
• Independently resolved live protected main: 7b81efd43c48439d7a4a5508f6d09ac15f141546.
• Fresh live comparison is diverged: the canonical branch is 33 commits ahead and 15 commits behind protected main, with merge base 4b2f5d30c1f1961e017a84aa1c785d3f1b1bb9af.
• GitHub currently reports this Draft non-mergeable.
• The branch is the canonical semantic owner, but it is not protected-main authority and is not integration-ready until deliberately converged onto current main.

No predecessor check, review, approval, synthetic merge result, remembered SHA, or documentation claim transfers to a replacement or changed head.

Canonical graph

The branch contains code-current equivalents of:

• PRD and TRD;
• root Architecture;
• ADR index and ADR-0001..0010;
• component, sequence, state, deployment, repository-authority, writer-lease, stale-convergence, incident, and recovery UML;
• conceptual/logical ERD and evidence/data model with actual persistence ownership and explicit no-invented-database boundaries;
• API/evidence contracts;
• Security and Threat Model;
• Test Strategy and exact coverage doctrine;
• Operability, Incident/RCA, and Recovery;
• Quality Attributes and Accessibility Acceptance;
• standalone/CWL Interoperability;
• privacy-safe Observability and Data Governance/privacy/retention;
• Roadmap;
• Release/Rollback/Provenance;
• Licensing/IP/NOTICE;
• Standards and APA 7 references;
• Acquisition Diligence and Traceability;
• documentation index, repository governance, CLAUDE context, CHANGELOG alignment, and executable documentation-fitness tests.

Whole-conversation documentation assessment

The verdict remains deliberately multi-axis:

1. Family coverage — sufficient on this active canonical owner.
2. Semantic depth — sufficient for the durable product, authority/evidence, standalone/MSA, automation, privacy, release, recovery, non-goal, and documentation-to-code decisions reviewed in the DiskSage conversation.
3. Protected-main authority — insufficient because the graph is not integrated.
4. Integration freshness — insufficient because this branch diverges from current protected main and must preserve newer protected-main source, workflow, dependency, security, governance, and changelog truth.

The assessment does not invent a central SQL database, measured SLO/RPO/RTO, remote telemetry, enterprise identity, release provenance success, legal rights, certification, or active-PR behavior as shipped.

Fresh 2026-08-11 reconciliation also keeps active implementation maturity separate from shipped truth: #150 remains the Draft Podman desktop evidence line, #156 remains the Draft exact-production-coverage line with a real broad coverage deficit, #167 is the current-main release-provenance successor with native exact-head workflows green but current-head semantic review still being reacquired, and #168 remains a Draft LLM structured-output parser fix whose central OpenCode review blocker is infrastructure/policy evidence rather than a source-backed finding. None of those active lines is protected-main product truth.

Scheduler-control and continuation governance

AGENTS.md on this branch records that a generic scheduled-task error, missed recurrence, user report that work remained, or prompt-repair request is control-plane evidence rather than completion or an invented source defect. Prompt/RCA/docs/one check/one merge/one slice has zero completion value by itself; same-invocation repository execution and a non-documentation handoff are required whenever safe work exists. Scheduler API unavailability blocks only scheduler mutation.

Test-first history

• RED c28bc81d71fa5b918917ee5c5c9648f6d1831fa2: require the canonical documentation families and critical authority/data-model/release markers while protected main lacks them.
• Product/technical spine through 14f559a43a7a6206119496552964c77a55ee0e0d.
• Diagrams/data/evidence through 82a8d153da8e83589ae0d732a2f84d9180c96ed2.
• Commercial/release governance through 8cbbc031861a58ce4dc5191e6a5eaac7b99c4e7f.
• Security/quality/operations through 3417de74659b0b7c2d23dea1c465cfe86c6ef7d5.
• ADR lifecycle through acf7a5f36229c0381d4bf9003a2c4ea9455c2b36 and subsequent ADR-0009/0010 convergence/documentation-authority additions.
• Current canonical head remains 4db841159b2f6b8c30bfc53d5aeb0365261d19c0; this PR-body reconciliation updates transient live evidence only and does not pretend the stale branch contents are protected-main-current.

Convergence requirement

Before this stale line is closed or integrated:

1. compare exact live protected main to this head and to the chosen successor;
2. preserve every valuable unique canonical file and semantic delta;
3. deliberately retain all 23 protected-main deltas since the branch merge base, including the removed obsolete repair workflow, current release/test workflow pins, tenant-authority/CSP/package-metadata behavior and tests, current dependency/lockfile state, and current source changes;
4. merge CHANGELOG.md semantically rather than overwriting either line; retain this branch's stronger AGENTS.md/SECURITY.md only where they do not contradict newer protected truth;
5. preserve the executable documentation regression without weakening production coverage scope;
6. classify every conflicting/obsolete delta with a technical rejection or supersession reason;
7. reacquire unchanged exact-head Test, Release, Security Scan, SAST, documentation, review, and live-policy evidence;
8. close this line only after the lineage map proves convergence.

Newer main or behind_by alone never proves supersession.

Required before merge

Keep Draft until a current-base converged head passes every applicable repository gate, has zero valid unresolved findings, satisfies actual live review/governance policy, and the documentation assessment can change protected-main authority from insufficient to current without overstating product/acquisition readiness.

[coderabbitai]

Important

Review skipped

Draft detected.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 73432dab-9eae-4a58-9e88-4aa4457c04d4

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[seonghobae]

@OpenCode review current head f9f11fb; fixed architecture documentation contract to match canonical Protected main wording