Skip to content

✨ Refresh v6 pilot and enable upstream ambient code-rain testing - #172

Closed
hanthor wants to merge 283 commits into
tuna/v6from
deploy/upstream-v6-matrix
Closed

hanthor wants to merge 283 commits into
tuna/v6from
deploy/upstream-v6-matrix

Conversation

@hanthor

@hanthor hanthor commented Oct 31, 2026 •

Copy link
Copy Markdown
Member

Purpose

Refresh the Spindle pilot onto upstream v6 6f6a5ca63 while preserving the proposal gate, broker access, freeze windows, proxy device-flow, spoke proof, omp ioscan and revert sheriff fork behavior.

Upstream v6 predates ambient Matrix mode, so this ports upstream hivecommons#15945, hivecommons#15946, hivecommons#15967, hivecommons#15982 (privacy fix), hivecommons#15977, hivecommons#16020, hivecommons#15985, and hivecommons#16046. All four effects (Rain, Cascade, Ghost coder, Terminal) included; existing v6 kiosk retained. The display config is named DashboardMatrixConfig internally to avoid collision with v6 Matrix chat notifications. Source remains default-off; Twitch overlay remains inactive.

Verification / rollout

Only local check: gofmt -l on hand-integrated Go files, no output. No local Go tests/build-all/vet or citation guards. CI is the verdict: dispatch full v2 Tests and build stamped amd64/arm64 artifacts in fork-only Tuna pilot binary workflow. Deployment is scoped to Reef and School Matrix testing, per the operator’s clarification; Hub remains unchanged. Keep previous binary and stopped-service data snapshots, explicitly allowlist verified-public watched repositories, deny unauthenticated line requests, and prove API lines against a genuine GitHub PR patch.

No protections or sentinel exemptions are relaxed. Deployment evidence will be appended once exercised.

Current deployment: repository-name fix on both spokes

Exact deployed commit: 637bae44f2ae9ac3c53f73e7557c71a9d4b657a7. These hives configure project.org: tuna-os plus short repository names, while Matrix requires owner/name. The sentinel passed a short display name to the observer, so every patch failed the allowlist. The observer now receives the resolved owner/name; privacy, authorship, allowlist and writeback gates are unchanged. Added a short-repository regression case.

  • Full fixed-head Go CI: https://github.com/tuna-os/hive/actions/runs/41233790530 — 27 passed, three skipped.
  • Fixed binary build: https://github.com/tuna-os/hive/actions/runs/41233789674 — passed. ARM64 artifact and live hash: b930b03e3d4e2d88baf9dc963cae995859d2ea5c659ef407e10b5b355b4ae5c6.
  • Reef fix canary started 2026-10-31 21:56:45 UTC; School started 22:36:41 UTC after the CI pass. Both running with zero restarts and stamped version 6-tuna-pilot-637bae44f2ae.
  • Matching UI is byte-identical to 71ce0ef56; both retain dashboard-71ce0ef56.html. Hub remains 230f745 and was not restarted.
  • Consistent fix snapshots/config/override backups: Reef 20261031T215350Z; School 20261031T223247Z. Immediate previous binary 71ce0ef56 retained.
  • Live capture and rendering verified. Nonempty feeds observed on School at 23:11 UTC and Reef at 23:18 UTC; both subsequently returned seq: 1200 with 1,200 buffered lines, HTTP 200 and Cache-Control: no-store.
  • A sample from each feed matched a genuine added line after whitespace trimming in public tuna-os/Tavern#323, src/url_security.py, head ac9349dba541c1c9297a4282565ce577ceabd7a9. Shared watches mean per-spoke buffers can overlap; this is not a deduplicated hub total.
  • Actual browser proof: Rain/Ghost on Reef, Cascade/Terminal on School; P-key pause/resume visibly worked. Anonymous requests, including forged X-Hive-Role: owner, returned 401 on both populated feeds.
  • Reef’s first cold start exhausted its App quota; captures recovered after reset with all guards intact. The current setup is polling-backed PR-patch playback, not live editor keystrokes.
  • 0d4a4d3e6 is a subsequent docs-only scope correction: Hub remains untouched and the Node-served HTML mount is documented. Runtime pin and successful Go-suite verdict remain the exact 637bae44f code/artifact.

Operator feedback: prefer a moving recent-code window with enough history for dense idle playback, prioritize new patches and distinguish replay from fresh activity. The renderer already retains/reuses a bounded buffer; larger-history/replay-indicator controls are a proposal, not delivered settings. Hub-wide aggregation remains deferred by the operator’s “fix first, then decide” selection.

Initial 71ce0ef56 rollout evidence (historical)

The inherited squash-attribution check flags imported upstream history because fork default v4 lacks already-landed upstream v6 commits. Preserve upstream authors/history; do not squash/rewrite upstream sign-offs. Keep this PR unmerged pending proper provenance-policy resolution; no branch protections or sentinel exemptions were relaxed.

clubanderson and others added 30 commits October 28, 2026 23:48
…contributor, not Sentinel alert (hivecommons#15256)

* fix: clanker PRs with a human changes-requested review show Returned to contributor, not Sentinel alert

A trusted human's CHANGES_REQUESTED review on the current head of a
clanker-authored PR is the human the sentinel/hold/workflow-path parks
were waiting for: RouteHuman and ClassifyStage now route it to
stage/072-returned-to-contributor (071 for a person's PR, 070 for the
hive's) ahead of 085, and the contributor lane reads reviews before it
settles on sentinel-blocked. A new head re-arms the park (per head, as
in hivecommons#15216).

Dashboard direction decisions are accepted on clanker PRs: the text is
recorded as a change request against the head, posted to the
contributor's hive, and the PR moves to 072; the 409 stays for a
person's PR only.

SPA: the 071/072 chevron keeps its routing word with the sentinel
finding as a suffix; Give direction names the contributor's hive.

Fixes hivecommons#15255

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* fix: honour a hand-placed stage/071/072 over the sentinel park for the current head

The reconciler logged added=[stage/085] removed=[stage/072-returned-to-
contributor] on hive#15159 after the owner swapped the labels by hand.
A trusted maintainer's return label the hive did not write is now read
from the label event once per head, recorded as a stage-label change
request against the head, and folded into the pass like a review, so
classification keeps 071/072 ahead of the sentinel/hold/workflow-path
parks; a new head re-arms the park. A standing change request in the
store now also folds into the stage detail when no reviewer verdict
was read for the head.

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* fix: drop unused openPullRequestHead wrapper

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

---------

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…only Prow for rejection replies (hivecommons#15275)

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
…e next rotation point (hivecommons#15270)

* 🐛 fix(kick-logs): archive kick log when the turn ends, not only at the next rotation point

A kick's log was archived only at a rotation point (next kick, restart,
graceful shutdown). A turn that ended lived only in tmux scrollback until
then; a hard container death in that window (exit 17, OOM — no SIGTERM)
lost the run, and past kicks showed nothing for the day.

Archive under reason turn-end when settleKickOutcomeLocked observes the
turn end. Scrollback and kickLogPending are left alone so the next
rotation point still takes the final snapshot and clears; that snapshot
supersedes the turn-end file (same LastKick stem) so each kick keeps one
archive and retention still counts kicks. A settled turn torn down later
is archived but no longer recorded as an interrupted turn.

Fixes hivecommons#15264

Signed-off-by: Andy Anderson <andy@clubanderson.com>

* 🌱 chore: move changelog fragment to repo-root changelog.d

Signed-off-by: Andy Anderson <andy@clubanderson.com>

---------

Signed-off-by: Andy Anderson <andy@clubanderson.com>
…; decide a post-decision re-park anew (hivecommons#15278)

A hive-authored PR whose issue another merged PR closed, or that a merged
PR declares it supersedes, is closed with ledger reason superseded-by-merge
before the stage pass routes the reviewer's requires_human (or a conflict)
to 085/070; the self-authored sweep asks the superseded rule before the
needs-rebase label block. A closing claim closed as not planned with no
merged closer also closes the PR.

The delegated-decision dedup now notices a park label applied after the
recorded decision and decides the new ask instead of posting the
decision-stalled comment.

Fixes hivecommons#15276

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
* 🐛 dashboard: merged toast shows the PR number

The merge banner rendered #${esc(info.number)} with a Number; the page's
esc() returns '' for any non-string, so the head read a bare "# merged".
Resolve the number as a string (field, nested pr record, GitHub URL, key),
never render a bare "#", and make the node harness's esc stub mirror the
real one so the case is covered.

Fixes hivecommons#15277

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* 🐛 Merged list: numbers and counts through String() too (same esc() non-string drop)

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

---------

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…5281)

A PR row returned to its author (071 / 072) now carries rework{state,
actor, requested_at, since, evidence, mode, ahead, head}: requested
(evidence queued when the fix sits on the clanker's hub queue),
acknowledged (PR updated_at moved after the request and the hive's own
nudge), in_progress (the clanker relay took the fix-pr task) or pushed
(head moved; 050 on the next sweep). Derived from the stage record, the
PR list and the contributor fix queue — no extra GitHub calls.

The waiting-on / adoption pill appends the state word, its tooltip and
the row tooltip say since when, and a state change at an unchanged stage
writes the row event line. fixqueue.Dispatch records when it left the
queue.

Fixes hivecommons#15280

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…mmons#15285)

A paused spoke (every agent paused, both repos in project.paused_repos)
still edited its pinned '## 🐝 Advisory Digest' comment at startup while
the auto-merge sweeps correctly logged repo-auto-merge-disabled. Repo
pause (hivecommons#6203) is documented as stopping all agent writes; the digest
path never consulted it.

- pkg/writeback: Deps.RepoPause lookup; the gate refuses every kind on a
  paused repo with reason repo-paused, checked before mode/own-item, and
  records the suppression. Gate.RepoPaused exposes the lookup for writers
  that are not a gate kind.
- pkg/config: WritebackGateDeps wires RepoPauseFor (by/reason) into it.
- pkg/github: PostAdvisoryDigest returns *RepoPausedError before any
  forge call, logging 'advisory digest skipped: repo paused' with by and
  reason and auditing advisory_digest_suppressed reason=repo-paused; no
  hash state is touched so the first cycle after resume posts.
  SweepDuplicatePRs and SweepClosedIssueStageLabels iterate activeRepos.
- cmd/hive: publishAdvisoryDigest treats RepoPausedError as a skip (not a
  post, not an error, no App-banner hooks) and paces the next attempt by
  the update interval; hard-suppress reconcile uses ActiveRepositories.
- docs: repo-pause.md, operator-reference.md, github-write-surface.md.

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…ide rows (hivecommons#15186)

Operator pressed Let hive decide on three 090 rows and the row showed a
green check, a Hive is deciding spinner and the unchanged striped Ready
for human merge pill at once; later the same rows read Approving for an
adopt-supersede, hive's copies appeared as unrelated sub-rows, closed
originals became blank numbered rows, a recorded decision left the row
parked in 085 under a stale needs-human label, and event lines spoke in
raw stage codes (PR at 060 reviewed) while 060 was drawn in the merge
gate column.

- One post-decision projection (stage_tracker_decided.go): deciding ->
  decided + effect pending -> landed / timed out. The check is the
  decision badge, never a celebration; pill, chevron and progress text
  name the recorded verdict (Approving only for approve); toasts lead
  with the outcome in one line.
- Preview before acting: POST .../decide-delegated/plan (dry run, no
  side effects) and a confirm popover Let hive do that / Approve & merge
  as-is / Cancel; chosen button recorded in the trail.
- Adoption pairs: dimmed Adopted -> #N original, adopted-from chip on
  the copy, closed original as one cached-title line.
- Any recorded decision lifts needs-human + 045/085 labels in the same
  write; decided rows never project into 085.
- Shared stage vocabulary (stage_tracker_vocabulary.go + SPA mirror)
  with a guard test; 060 reviewed drawn in CI / Review as Review passed
  · <lane reason>; merge gate column starts at 080.
- The author's clanker, not the author's hive; local relative times in
  the no-effect popover; in-tracker maintainer merge when available.
- openapi.json + api-reference row for the plan route; docs; changelog.

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…5290)

* 🐛 dashboard: merge toast renders above the chat button

The merged-toast stack (#merge-banners) and the .cue-flight glyph are
appended to <body>, but their layer tokens --z-merge-banner /
--z-cue-flight were declared on #activity-monitor-section. Outside the
section they resolved to nothing, so the stack had z-index:auto and the
chat FAB (z-index 9000) painted over the lowest toast's close × and
truncated its text. Hoist both tokens to :root; add a static guard.

Fixes hivecommons#15288

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* 🐛 dashboard: avoid a literal body tag in the merge-banner layer comment

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

---------

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…ed conflicting PRs; dismiss self-inflicted changes_requested (hivecommons#15104)

* 🐛 Activity Monitor: Rebase, then merge as maintainer; autonomous rebase-resolve with loop guard; approve on a dirty PR queues the rebase; dismiss self-inflicted changes_requested

Squashed and rebased onto v5 after hivecommons#15087 and hivecommons#15074 landed.

Fixes hivecommons#15103

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* test: fix fixtures after the v5 rebase (mock fields, failingEntry rebase fields, import ratchet, window shim)

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* test: lift pkg/rebasemerge coverage above its floor; drop an ineffectual assignment

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* test: compare the shared singletons through named values (staticcheck SA4000)

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* security: rebase-then-merge fails closed on a head the hive did not produce; document the new API shapes

The approval is bound to the head the operator read. A contributor PR whose
head moves after the click now counts as rebased only when GitHub reports
the hive (App bot or pane identity) as the commit's author or committer;
any other head settles the follow-through as failed (foreign-head) with
nothing merged, a note on the PR, and 'Approve the new head & merge' as the
way out. api-reference.md gains the 202 rebase-queued answer, the rebase
response field, and the rebase_merge / rebase row fields.

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* merge train: hand a dirty member to the fixer's rebase-resolve episode; the exhausted row's rebase-then-merge option goes live

v5 landed the conflict-aware merge train (hivecommons#15157) with a RebaseResolver hook
reserved for this PR and a duplicate rebaseMergeFollowThroughLabel. The hook
now opens the shared rebase episode (requested_by: merge-train); the
mergeNowWired flag and its disabled reason go, since the option it guarded
is wired here.

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* test: approve toast wording follows hivecommons#15213

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* rebase onto v5: drop duplicate nonEmptyString, lift STAGE_TRACKER_ICONS for the history badge in the rebase-then-merge row test

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* writeback ratchet: baseline the two operator-solicited comments of the rebase-then-merge follow-through

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* automerge: superseded question precedes the label block (hivecommons#15278) while a needs-rebase label on a merely-behind head stays stale

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

---------

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
)

Document the never-nudged 071 pickup limitation and distinguish activity acknowledgement from commit evidence. Re-review still proceeds on the next stage sweep.

Hive-Run: hivecommons#15286
Hive-Plan: document-nudge-baseline-limitation
Hive-Spec: hivecommons#15286#document-limitation

Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>
Co-authored-by: thatguywob <thatguywob@users.noreply.github.com>
… align their populations (hivecommons#15302)

The Overview pill prints the open set as items (33 issues / 13 PRs); the
Hive activity monitor pill prints the same set as rows (every issue, plus
each PR not drawn under a listed issue). Nothing said so, so 33/13 beside
37 read as a disagreement. Both tooltips and the activity-monitor header
now name the unit and spell out the row make-up
(37 rows (33 issues + 4 PRs on their own row) · 9 PRs shown on their
issue's row). Docs corrected; node-harness tests cover the derivation.

Fixes hivecommons#15295

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
… un-park/re-park loop, name the missing App permission (hivecommons#15134)

* Model capability gaps as first-class blockers so delegated decisions stop re-parking loops

An issue the scanner parks because the GitHub App cannot push
.github/workflows/* (no workflows permission) is a capability gap, not a
question: a direction decision cannot clear it, and "Let hive decide"
un-parked hivecommons#14976 only for the sweep to re-apply the same
needs-human text an hour later.

- pkg/blocker: structured blocker kinds, the <!-- hive:blocker kind=… -->
  marker, prose inference, and a persisted ledger with a delegated-unpark
  loop guard and a cleared-at staleness guard.
- Record blockers where comments are read: issue_comment webhook, task-list
  sweep (adds the marker to the remainder comment), gh-wrapper (adds the
  marker to a needs-human comment that names a missing capability).
- Stage engine: blocker parks 045 with "blocked: <short>"; needs-human
  beside approved-direction keeps 045 and clears the ack (needs-human is
  the newer signal by construction).
- Dashboard: decision_needed.blocker/options populated with real handlers
  (App settings link, Re-check permissions, close as partially addressed,
  wait; patch hand-off disabled with its reason), delegated decision
  short-circuits to no-effect with the glossary 'blocker' reason, the row
  pill reads "Blocked · App lacks workflows permission", ⚙ → Security and
  /api/status.capabilities name missing permissions and blocked counts.
- Tests: Go (blocker, ledger, webhook, sweep marker, stage facts/derive,
  dashboard handlers, glossary) and node fixtures (pill, options, popover,
  capability text); gh-wrapper bash tests.
- Docs: labels-and-control-signals (blocker marker + kinds + reconcile
  rule), dashboard-sections, operator-reference (App permissions table),
  api-reference; changelog fragment.

Fixes hivecommons#15099

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* Blocker options carry a plain-words consequence for the popover

Follows hivecommons#15087's option consequences so a blocked row's ways out say
what each one does.

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* docs: stable anchor for the Blockers section

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* fix: close-partial response carries the result by pointer

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* fix: pkg/github blocker import alias, ledger loop-guard timing, test call sites, import ratchet

- pkg/github already has a test helper named blocker; import pkg/blocker
  as itemblocker there.
- Ledger.Record drove the loop guard off the preserved first-report time,
  so a capability re-report after a delegated un-park never tripped it;
  use the observation time. Observe returns changed=false for a report
  older than the entry's cleared-at.
- stageTrackerDecisionNeededWith test calls pass the clock.
- Allow pkg/dashboard → pkg/blocker in the import ratchet (the blocker
  ledger is the seam the tracker, delegate and status read).

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* fix: keep the import allowlist sorted

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* Fix recheck/held tests, document blocker routes in openapi.json, drop duplicate JS const

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* test: cover blocker kind words, ReasonSentence caps and ledger edge cases; resync api-reference citations after rebase

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* fix: mark the close-partial blocker writes as operator-solicited for the write-back ratchet

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* docs: resync api-reference citations after rebase onto v5

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

---------

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…age directive (hivecommons#15293)

* [sec-check] fix: read sandbox publish.verify from config only, ignore kick-message directive

The kick message interpolates issue/PR titles, knowledge-base fact bodies
and delegated prompt text, none of which the hive authors. Taking the
publish.verify command list from a <!-- hive:publish.verify --> block in that
message let two crafted work-list title lines choose the shell commands run
over the agent's tree before its commits are pushed.

Recognise the directive for logging but never run it; config
(sandbox.publish.verify) is the only source of verify commands.

Signed-off-by: sec-check <280983584+hivecommons-hive[bot]@users.noreply.github.com>

* 📝 changelog: drop kick-template directive from publish.verify fragment

The earlier unreleased fragment still advertised the <!-- hive:publish.verify -->
template block; verify commands now come from config only.

Hive-Fix-Attempt: 1/6

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>

---------

Signed-off-by: sec-check <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: sec-check <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
…xt sweep (hivecommons#15306)

check_run/check_suite/status failure webhooks now resolve to open PRs in
tracked repos, record a CI-failure signal (dedup per head SHA + check),
and immediately kick the owning agent through the FIX-BEFORE-NEW red-CI
path or nudge the contributor lane for a targeted automerge pass. The
sweep still merges the signal store as fallback. Guarded by
governor.ci_failure_webhook_kick / governor.ci_failure_kick_cooldown_s,
repo pause, agent pause, and governor eligibility.

Fixes hivecommons#15291

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…ge' (hivecommons#15253)

The Let-hive-decide result toast (and the /hive decision comment and the
stalled follow-up) used PR wording for issues: on hive's own spoke,
issue hivecommons#15021 read "Hive decided approved for merge at
2026-10-28T20:41:22Z, but it hasn't taken effect: the item is still
waiting on a person" — nothing merges on an issue, the headline said
"decided" twice, the stamp was raw ISO, and the real cause (the
stage/045-needs-decision park label outlived the approved-direction it
recorded) was neither named nor fixable from the toast.

- pkg/delegatedecision: GlossaryEntry gains IssuePlain / IssueDetail;
  PlainOn / DetailOn / VerdictPlainOn / ActionPlainOn select by item
  kind (approve → PR "approved for merge", issue "approved to proceed —
  an agent may claim it"; direction, close-superseded and the hold
  reason likewise). New reason slug park-label-stuck. GlossaryJSON
  mirrors the issue words (issue_plain / issue_detail).
- dashboard: delegatedStuckWhy detects the stuck 045 park label on an
  approved issue; delegatedStuckOptions puts "Lift the park (remove
  stage/045-needs-decision)" (decide:approve → RecordOperatorIssueDecision,
  the audited path that removes the label and tells the stage engine)
  and "Re-evaluate now" first. The no-effect outcome no longer repeats
  "Hive decided".
- SPA: STAGE_TRACKER_DECISION_GLOSSARY regenerated;
  stageTrackerPlainWords(slug, detail, itemKind); no-effect headline
  word is "no effect yet"; stageTrackerStampWords rewrites RFC3339
  stamps into the operator's clock via formatDashboardDateTime.

Fixes hivecommons#15252

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…ns#15236)

* 🐛 fix: prefer covering PR progress in tracker issue messages

Signed-off-by: Andy Anderson <andy@clubanderson.com>

* 🐛 fix: preserve independent issue actors outside PR narration

Signed-off-by: Andy Anderson <andy@clubanderson.com>

* 🐛 test: expect PR progress instead of stale scanner actor

Signed-off-by: Andy Anderson <andy@clubanderson.com>

---------

Signed-off-by: Andy Anderson <andy@clubanderson.com>
…ing them (hivecommons#15294)

* 🐛 fix(writeback): gate adoption close-first writes instead of baselining them

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>

* 🐛 fix(writeback): warn on policy-refused close-first reopen; document the gate in the close-first prose

Review follow-ups for hivecommons#15294:

- reopenSupersededOriginal logs a non-pause gate refusal at WARN (a pause
  stays at info): the contributor_adopt policy moved after the hive closed
  the PR, so the reopen that would undo that close no longer clears the
  gate and the PR stays shut until a person acts. New test
  TestAdoptionCloseFirstReopenPolicyRefusalWarns.
- operator-reference.md and labels-and-control-signals.md close-first prose
  now name the contributor_adopt gate and the refused outcome, linking the
  write-back table row.

Hive-Fix-Attempt: 1/6
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>

---------

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Explain the shared open-items population in the Overview tooltip and show the activity monitor row make-up in its header and tooltip.



Hive-Run: hivecommons#15295
Hive-Plan: dashboard-counter-units
Hive-Spec: hivecommons#15295#fix

Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>
Co-authored-by: thatguywob <thatguywob@users.noreply.github.com>
…ivecommons#15301)

* 🐛 Let hive decide: close already-fixed PRs instead of adopting them

Delegated decisions on contributor PRs now walk a ladder before the
contributor-offline adopt-supersede rule: already-fixed (issue closed by a
merged PR, or a merged PR declares supersession) → close as superseded
with a note naming the merging PR; mergeable-as-is → merge gate;
rebase-then-merge (recorded, lands with hivecommons#15104); return-to-contributor
when the author's clanker or hive is connected or picked the rework up;
adopt-supersede only last, never for owner/human PRs. Each rung logs why
the previous ones did not apply; the toast says why; the decide API
exposes decision.ladder {decision, reason, ruled_out}. Wording names the
author's clanker vs hive truthfully.

Fixes hivecommons#15289

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* 🌱 dashboard: stage_tracker_linked_progress_test compiles against the open-vouch signature

v5 landed hivecommons#15236 and hivecommons#15253 side by side; the former's test still called
stageTrackerResolveLinkedPRs with one argument.

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

---------

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
)

* 🐛 fix: require clanker label for dashboard directions

Reject author-controlled PR-body markers as standalone evidence for the dashboard contributor direction lane. Cover all three marker types with no classification and with author/human, asserting no stored decision or GitHub/stage writes.

Hive-Run: hivecommons#15268
Hive-Plan: dashboard-direction-classification
Hive-Spec: hivecommons#15268#body-marker-trust
Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>

* 🐛 fix: honor configured author labels in dashboard directions

Resolve dashboard decision authorship through the live governor.labels.author names, wired through shared client configuration so rebuilt clients retain the policy. Cover renamed and partial configurations and reject stale default labels without writes. Clarify that sweep body-marker classification remains unchanged.

Hive-Run: hivecommons#15268
Hive-Plan: dashboard-direction-classification
Hive-Spec: hivecommons#15268#body-marker-trust
Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>

* 📖 docs: reanchor issue ranking citation after client change

The CI citation guard reported drift; its fixer could not resolve AgeMinutes as a top-level declaration. Reanchor the range to RankActionableIssues and its age comparison.

Hive-Run: hivecommons#15268
Hive-Plan: dashboard-direction-classification
Hive-Spec: hivecommons#15268#body-marker-trust
Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>

---------

Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>
Co-authored-by: thatguywob <thatguywob@users.noreply.github.com>
)

Clear blocker entries after successful issue closes and on closed issue webhook payloads. Ignore blocker reports embedded in comments on closed issues, and cover repository name aliases and failed or gated closes.

Hive-Run: hivecommons#15155
Hive-Plan: hivecommons#15155
Hive-Spec: hivecommons#15155#closed-issue-blocker-lifecycle

Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>
Co-authored-by: thatguywob <thatguywob@users.noreply.github.com>
src/docs/api-reference.md becomes an index page; each former section is
now its own page under src/docs/api-reference/<area>.md (15 areas, 620
routes, content moved verbatim with relative links re-rooted). The index
keeps every former ## heading so api-reference.md#<section> anchors still
resolve.

check-api-reference-citations.sh walks the index plus api-reference/*.md
by default (files or directories may still be passed), labels drift by
page, and --fix rewrites only the page that drifted. check-docs-citations.py
--fix passes the directory through; the self-test gains a case for the
directory walk and per-page --fix. Workflow messages, AGENTS.md and the
README/documentation map point at the new layout.

Fixes hivecommons#15299

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…5300)

* 🌱 dashboard: generate index.html from per-area sources

static/index.html is now a generated artifact. Its sources live in
static-src/ (head, css/<area>.css, glue, js/<area>.js, html/<area>.html)
listed in manifest.txt and concatenated verbatim by
webstatic/gen/concat.go (go generate; make dashboard-gen). The committed
artifact is byte-identical to before, so //go:embed, the CSP style
hashes, the node harnesses and every test reading index.html are
unchanged. TestIndexHTMLIsGenerated fails when artifact and sources
drift; TestIndexManifestEveryFileListedAndUsed keeps the manifest honest.

Docs: src/docs/dashboard-source-layout.md (layout, area list, conflict
recipe: take base's index.html, re-split delta, regenerate).
Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* webstatic/gen: make concat testable (run), cover manifest/concat error paths

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

---------

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…mons#15330)

* chore(crux): move tmux session accessors out of manager.go

SessionMissing and TmuxSession are tmux-session queries; they belong with
the rest of the tmux boundary in manager_tmux.go. hivecommons#15270 grew
agent/manager.go to 1696 lines, over the 1689-line baseline enforced by
crux's Go file-size ratchet, so every crux-publish sync since then fails
that check on crux main. Pure move, no behaviour change.

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>

* changelog: fragment for hivecommons#15330

Generated by changelog-fragment-guard from the PR title (hivecommons#5675).
Edit the wording freely, or replace this file with the no-changelog
label if the change is not user-visible.

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Signed-off-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>

---------

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Signed-off-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Automated pin bump (hivecommons#8419). Digests recomputed from the downloaded
release artifacts by src/scripts/cli-pin-bump.sh:

  VERSION=1.57.0
  SHA256_AMD64=10eb96cb6a159b6e1d38eeedbef6ab64cd28168aad91948f1fa5c34eefdf98e3
  SHA256_ARM64=c32605acfd560aeb2adbf17ec359cc0217e28022a7e647b35a6af5dcfca91749

Signed-off-by: hive-release[bot] <hive-release@hive.kubestellar.io>
Signed-off-by: clubanderson <407614+clubanderson@users.noreply.github.com>
Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: hive-release[bot] <hive-release@hive.kubestellar.io>
Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
* 📖 docs: make guides approachable and clarify current policy

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* 📖 docs: reconcile stop-label and review override guidance

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

* 📖 docs: keep cost help aligned with dashboard descriptions

Signed-off-by: Andrew Anderson <andy@clubanderson.com>

---------

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
…mmons#15339)

* 🐛 fix: stop retrying shared-ci reruns the App may not perform

A 403 'Resource not accessible by integration' on rerun-failed-jobs is a permanent installation permission gap. One PR with N red workflows emitted N identical WARNs per pass. Stop the pass on the first refusal, remember it per repo for 1h, warn once, and classify the new message in fleetreport.

Closes hivecommons#15336

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>

* changelog: fragment for hivecommons#15339

Generated by changelog-fragment-guard from the PR title (hivecommons#5675).
Edit the wording freely, or replace this file with the no-changelog
label if the change is not user-visible.

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Signed-off-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>

* changelog: drop malformed duplicate fragment

changelog-fragment-guard requires a '- ' entry bullet; the auto-added
fixed-15339.md already carries the entry.

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>

---------

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Signed-off-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
* 🌱 chore: remove unreachable MITM cancellation guard

Requests parsed by http.ReadRequest have no cancellation context attached in the MITM path, so this guard cannot fire. Retain caller-cancellation handling in the HTTP translator handler.

Hive-Run: hivecommons#15089
Hive-Plan: MITM cancellation dead-code cleanup
Hive-Spec: hivecommons#15089#unreachable-mitm-guard
Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>

* 🌱 chore: add changelog fragment for MITM cleanup

Record the MITM dead-code cleanup so changelog-fragment-guard passes without requiring label permissions.

Hive-Run: hivecommons#15089
Hive-Plan: MITM cancellation dead-code cleanup
Hive-Spec: hivecommons#15089#unreachable-mitm-guard
Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>

* 🌱 ci: allow full image build time for overlayfs guard

The guard exhausted its ten-minute job budget while the optional Bob CLI download was retrying, before reaching its containerd assertions. Match the existing docker job budget because both build the full src/Dockerfile image.

Hive-Run: hivecommons#15089
Hive-Plan: MITM cancellation dead-code cleanup
Hive-Spec: hivecommons#15089#unreachable-mitm-guard
Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>

* 🌱 chore: keep MITM cleanup scoped to proxy dead code

Drop the unrelated overlayfs CI timeout increase as requested by maintainers. Retain only the unreachable MITM cancellation guard removal and its changelog fragment in the PR diff.

Hive-Run: hivecommons#15089
Hive-Plan: MITM cancellation dead-code cleanup
Hive-Spec: hivecommons#15089#unreachable-mitm-guard
Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>

---------

Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>
Co-authored-by: thatguywob <thatguywob@users.noreply.github.com>
…#15334)

* 🐛 fix: classify inference caller identification refusals

Report repeated translator identity refusals as a medium-severity inference gateway condition while preserving the fail-closed authentication guard.

Hive-Run: hivecommons#15333
Hive-Plan: classify-inference-caller-refusals
Hive-Spec: issue-15333#runtime-error-attribution
Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>

* 🐛 fix: cover unidentified inference caller threshold

Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>

---------

Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>
Co-authored-by: thatguywob <thatguywob@users.noreply.github.com>
@tuna-reef-hive tuna-reef-hive Bot added stage/071-needs-author Human-authored PR: CI red, changes requested or conflicting; diagnosis posted, the author must act stage/050-review-pending Review swarm requested; awaiting a verdict for the current head and removed stage/050-review-pending Review swarm requested; awaiting a verdict for the current head stage/071-needs-author Human-authored PR: CI red, changes requested or conflicting; diagnosis posted, the author must act labels Nov 2, 2026
@tuna-reef-hive

tuna-reef-hive Bot commented Nov 6, 2026

Copy link
Copy Markdown
006073d

ℹ️ Sentinel notice — informational; author is trusted (owner), merge is not blocked

Hive flagged this PR (author @hanthor, head 006073db9686) because it matches behaviors that can override security controls, escalate privileges or damage the codebase. This is a heuristic, not an accusation — a maintainer should confirm the change is intended before it merges.

  • sensitive_path — changes 55 sensitive path(s) (Touches a sensitive path (OWNERS, workflows, policies, hive config, security docs, …))
    • .github/release-lines.yml
    • .github/workflows/changelog-fragment-guard.yml
    • .github/workflows/coverage-hourly.yml
    • .github/workflows/crux-file-size-ratchet.yml
    • .github/workflows/dco-post-merge.yml
    • .github/workflows/docs-link-check.yml
    • .github/workflows/kustomize-render.yml
    • .github/workflows/prometheusrule-check.yml
    • .github/workflows/stream-image.yml
    • .github/workflows/tuna-pilot-binary.yml
    • Makefile
    • bin/gh-wrapper.sh
    • bin/gh-wrapper.test.sh
    • src/Dockerfile
    • src/Dockerfile.contributor
    • src/crux/proxy/rules.go
    • src/deploy/README.md
    • src/deploy/k8s/deployment.yaml
    • src/deploy/k8s/secret.yaml
    • src/deploy/kustomize/components/monitoring/README.md
    • src/deploy/kustomize/components/monitoring/hive-metrics.json
    • src/deploy/kustomize/components/monitoring/kustomization.yaml
    • src/deploy/kustomize/components/monitoring/patch-deployment.yaml
    • src/deploy/kustomize/components/monitoring/prometheusrule.yaml
    • src/deploy/kustomize/components/monitoring/servicemonitor.yaml
    • src/deploy/kustomize/components/monitoring/tests/prometheusrule_test.yaml
    • src/deploy/stream/.gitignore
    • src/deploy/stream/Dockerfile
    • src/deploy/stream/README.md
    • src/deploy/stream/authproxy/main.go
    • src/deploy/stream/authproxy/main_test.go
    • src/deploy/stream/docker-compose.yaml
    • src/deploy/stream/entrypoint.sh
    • src/deploy/stream/healthcheck.sh
    • src/deploy/stream/k8s/configmap.yaml
    • src/deploy/stream/k8s/deployment.yaml
    • src/deploy/stream/k8s/kustomization.yaml
    • src/deploy/stream/k8s/namespace.yaml
    • src/deploy/stream/k8s/networkpolicy.yaml
    • src/deploy/stream/k8s/secret.yaml
    • src/deploy/stream/overlays/hive-hosted-matrix/kustomization.yaml
    • src/deploy/stream/overlays/hive-hosted/kustomization.yaml
    • src/go.mod
    • src/pkg/policies/ci_hotspot_test.go
    • src/pkg/policies/dedupe_policy_test.go
    • src/pkg/policies/defaults/ci-maintainer-advisory.md
    • src/pkg/policies/defaults/ci-maintainer-full.md
    • src/pkg/policies/defaults/ci-maintainer-holdgated.md
    • src/pkg/policies/defaults/ci-maintainer.md
    • src/pkg/policies/defaults/dedupe-issues.md
    • src/policies/ci-maintainer-advisory.md
    • src/policies/ci-maintainer-full.md
    • src/policies/ci-maintainer-holdgated.md
    • src/policies/ci-maintainer.md
    • src/policies/dedupe-issues.md
  • permission_escalation — widens workflow permissions or uses pull_request_target (Widens GitHub Actions permissions or switches to pull_request_target)
    • .github/workflows/stream-image.yml
  • ci_gate_weakening — CI configuration is deleted, softened or has gate checks removed (Disables or softens CI gates (continue-on-error, || true, removed checks))
    • .github/workflows/stream-image.yml
  • security_policy_edit — edits security policy, scanner config, rulesets or hooks (Edits SECURITY.md, dependabot/CodeQL config, rulesets or git hooks)
    • security-insights.yml
  • remote_code_execution — adds a remote-fetch-and-execute, decoded payload, reverse shell or embedded credential (Adds curl|sh, eval of encoded payloads, reverse shells or embedded private keys)
    • src/deploy/stream/Dockerfile
    • src/pkg/dashboard/job_terminal_test.go
    • src/pkg/dashboard/matrix_source_test.go
    • src/pkg/review/security_finding.go

Hive did not add the sentinel-alert label because this author is trusted. The finding remains recorded for audit and the dashboard Security tab; set sentinel.trusted_authors_block: true to require the blocking label for trusted authors too.

@tuna-reef-hive tuna-reef-hive Bot added stage/stalled In its current stage longer than expected stage/050-review-pending Review swarm requested; awaiting a verdict for the current head stage/071-needs-author Human-authored PR: CI red, changes requested or conflicting; diagnosis posted, the author must act and removed stage/stalled In its current stage longer than expected stage/071-needs-author Human-authored PR: CI red, changes requested or conflicting; diagnosis posted, the author must act stage/050-review-pending Review swarm requested; awaiting a verdict for the current head labels Nov 7, 2026
@hanthor

hanthor commented Nov 9, 2026

Copy link
Copy Markdown
Member Author

Maintainer decision: HOLD at needs-author / needs-human; do not promote or merge this head. The current squash sign-off check fails, the PR remains unstable, and the diff touches many privileged workflow/security-policy paths, including stream-image permissions and gate behavior. First restore a passing DCO/sign-off check; then obtain independent review of the workflow permission and publication changes. The trusted-author heuristic is informational, not an approval.

@tuna-reef-hive tuna-reef-hive Bot added the stage/stalled In its current stage longer than expected label Nov 10, 2026
hanthor added a commit that referenced this pull request Nov 13, 2026
Fork-only (tuna-os). Builds linux/amd64 and linux/arm64 hive binaries from
the exact tuna/v6 commit, stamps them with the commit and uploads them as a
30-day workflow artifact with SHA256SUMS for the Spindle Compose pilot.

Least privilege: contents: read only, no secrets, persist-credentials: false,
all actions pinned by SHA; nothing is published (no release, image, tag or
package). Classified as pinned to tuna/v6 in .github/release-lines.yml so the
release-line guard stays green.

Split out of #172 (originally 206e36a and ce45704); the obsolete
deploy/upstream-v6-matrix trigger and unused pull_request refs are dropped.

Signed-off-by: James Reilly <jreilly1821@gmail.com>
hanthor added a commit that referenced this pull request Nov 13, 2026
…iteback gate

Upstream's writeback containment ratchet (pkg/writeback TestDirectWriteRatchet)
refuses ungated direct writes. The fork's revert sheriff now consults the
shared ci_rerun writeback policy before rerunning a failing job and again
before opening a held revert, so operator pauses and per-repo write
restrictions apply to it too.

Ported from #172 (71ce0ef) without its Matrix kiosk test edits, which the
upstream refresh supersedes.

Signed-off-by: James Reilly <jreilly1821@gmail.com>
@hanthor

hanthor commented Nov 13, 2026

Copy link
Copy Markdown
Member Author

Decision (for @hanthor): close and split, as the owner hold asked: upstream refresh separate from the fork-only and security-sensitive changes, with original attribution kept.

Where each part of this PR went:

The Reef/School pilot keeps running its pinned 637bae44f binary. Closing this PR redeploys nothing. Once #179 lands, #178 builds stamped tuna/v6 binaries for any future rollout.

@hanthor hanthor closed this Nov 13, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

author/human A person pushed this PR themselves dco All commits are DCO signed-off needs-human role/trusted Author is trusted for unattended merge (trusted-author role or registry/claim tier in require_tiers) stage/stalled In its current stage longer than expected stage/071-needs-author Human-authored PR: CI red, changes requested or conflicting; diagnosis posted, the author must act

Projects

None yet

Development

Successfully merging this pull request may close these issues.

8 participants