Skip to content

🌱 Top up v6 with the latest v5 snapshot - #16378

Merged
clubanderson merged 307 commits into
v6from
sync/v5-to-v6-20261102
Nov 3, 2026
Merged

clubanderson merged 307 commits into
v6from
sync/v5-to-v6-20261102

Conversation

@clubanderson

@clubanderson clubanderson commented Nov 3, 2026 •

Copy link
Copy Markdown
Member

Summary

Ancestry-preserving v5 to v6 top-up. The final bounded incoming v5 snapshot is 03b781e3bf3fc11c9600b89b78038db5573956d8, frozen at 2026-11-03T16:00:38Z. The refreshed destination is d2dc0f66485f5766f34de0f9b05b04d0d4fb1107.

The refresh includes the 38 incoming commits through 6398b61d008f8ef99d817e749193c34ebd421c8c, plus the three subsequently landed fixes: refs/status-card notification deduplication (#16385), gated normal-mode stage notes (#16396), and canonical verified pin-commit provenance (#16400). Hive-source durable-lane changes and concurrent CI/Review activity lanes are included; this does not assert that the standalone Crux v0.2.0 release contains the newer lane APIs or that all consumers have adopted them. The weaker unmerged relay proposal #16398 and superseded #16394 are not imported.

True merge commits a4df37afadda771c3f6e20a10d621759544984e9, 63cfbb0b38b7adc61bc59febe6891777cd4c9571, and 66143614dbed95c6239f225719f4ec7e374f947b retain all prior top-up history, original source snapshot 660e5d1ccbf22fb590b38221438f56a2f67c142d, and the current v6 destination TIP-notice change. v6-specific configuration, routing, serialized merge behavior, approval fences, coverage floors, generated workflows/dashboard assets, and the exact combined scripts-tree pin are retained.

Review and admission

Independent precommit review cleared the large source integration. Parent review checked the final canonical proof, destination merge deltas, and attributable CI repairs. Genuine provider admission and required head-bound review/approval remain required on the published head 670776c942dbed8c3e9d82cc5593599071675e67; results on old heads do not substitute. No local Go verdicts were run, as required by repository policy.

DCO-signed repair 670776c942dbed8c3e9d82cc5593599071675e67 corrects citation drift, imports the byte-bound canonical CI NOTICE, restores original v6 OMP splash matching in the extracted leaf, and fixes strict legacy-status fixtures without changing production admission. The complete latest exact-head API snapshot has 76 successful checks, 18 skipped checks, and only two failed checks: DCO and its dependent native-queue-policy. Source build, test, lint, and coverage failures are repaired. Tide still awaits actual admission/approval.

The current genuine DCO App1861 check 126058962145 is UNKNOWN because REST returned 250 commits and complete-list GraphQL fallback did not finish. This is not a proven missing-signoff finding. No DCO substitution, permission expansion, empty commit, protection change or admission bypass is authorized. See #16404 for the owner-directed provider investigation.

Merge strategy

MERGE COMMIT ONLY. Never squash or rebase this forward merge. No production settings, integration execution enrollment, or deployment is activated by this PR.

clubanderson and others added 30 commits October 31, 2026 02:47
…mode (#15919)

In quiet notification mode the evidence pointer, the unblocked note, the
review backlog summary, the contribute-hub claim marker and relayed agent
status notes (awaiting-ci, parking, already fixed) become sections of the
single edited <!-- hive:status --> comment. The unblocked note is
idempotent per cause in either mode and skipped when the same merge closes
the issue; close-on-merge and the issue-stage lane post one closing
comment, not both. The client logs created/edited/suppressed counts hourly.

Closes #15918

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…15915)

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
…-hop tolerance (#15841)

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
* ✨ feat: fence integration promotion with exact-tested destination CAS

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

* πŸ› fix: require synced journal execution for integration promotion

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

* πŸ› fix: atomically fence candidate author head during promotion

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

* πŸ› fix: bind promotion to the resolved repository scope

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

---------

Signed-off-by: Andy Anderson <andy@clubanderson.com>
Co-authored-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Restore the approved historical compact preparation train pill and side rail on actual tracker rows, preserving live comments, exact-head uncertainty, stages, sorts and linked trees.

Independent code-review findings and actual first-head CI regressions were repaired. Final immutable head ec2efa0 passed all applicable CI, including dashboard behavioral regressions and complete image matrix. No rollout or configuration change.

Signed-off-by: Andy Anderson <andy@clubanderson.com>
…it repo allowlist) (#15945)

* ✨ feat: add opt-in dashboard.matrix config block (default off)

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

* 🌱 test: register dashboard.matrix in the settings-parity allowlist and runtime golden

The new default-off dashboard.matrix block has no Settings control yet
(nothing reads it), so the parity test needs an allowlist entry, and the
zero-value struct marshals as an empty object in the agent-runtime golden.

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>
…rivate-repo gates (#15947)

* ✨ feat: add opt-in rich RFC notify config with designated-announcer and private-repo gates

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

* fix: retry RFC notify events when repo visibility is unresolved

Hive-Fix-Attempt: 1/6

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

* 🌱 test: allowlist the rich RFC notify keys in the settings-parity test

The five new notifications.discord.rfc_* keys are operator yaml-only
gates with no Settings control yet, so the parity test needs entries.

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: architect <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>
* ✨ feat: bind candidate producer identity to trusted CI jobs

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

* ✨ feat: materialize controlled integration prefixes with isolated CI receipts

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

* ✨ feat: observe exact candidate and materializer run artifact provenance

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

* πŸ› fix: repair materializer contract assertion syntax from CI

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

---------

Signed-off-by: Andy Anderson <andy@clubanderson.com>
…etch (#15946)

* ✨ feat: add bounded optional PatchObserver to the sentinel PR-patch fetch

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

* πŸ§ͺ test: bump head SHA before incomplete-list sweep in PatchObserver fetch-failure test

The sweep records the head as seen before fetching files, so re-sweeping
the same head after a 502 was skipped instead of re-fetched and the
incomplete-list assertion failed every run.

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>
…keeps failing (#15953)

* πŸ› fix: reconcile effect-ledger ops stuck in Unknown

A label mutation whose write ended Unknown (or Planned) is fenced until it
is reconciled. reconcileLabelMutation reconciles it from one Issues.Get,
but when that read failed it returned the refusal unconditionally, with no
age bound. A read that keeps failing (item transferred or deleted, a
persistent API error) therefore pinned the op Unknown forever, refusing
every retry and surfacing as effect_ledger_op_stuck.

Every other reconciling caller goes through effects.ExecuteReconciling,
which expires an unresolved op to not-applied after DefaultUnresolvedTTL.
Apply the same bound here: a young op keeps the refusal, an op unresolved
past the TTL is expired and the label write replays. Adding or removing a
label is idempotent, so the replay is safe.

Closes #15944

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

* πŸ“ docs: refresh RankActionableIssues citation in scale-envelope

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

---------

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Signed-off-by: ci-maintainer <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>
…ode (#15948)

* ✨ feat: describe Activity Monitor filters in am.* URL params and add attended display mode

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

* πŸ› fix: keep Esc in display mode from leaving while a popover or dialog is open

The display-mode Esc listener only exempted the settings overlay, so Esc on
an open merged/queue popover or hive dialog reloaded out of display mode and
dropped session-only am.* edits. The listener now skips whenever any tracker
popover, settings/labels dialog or hive dialog is open, and the merged and
queue popover closers claim Esc with preventDefault.

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>

* πŸ› fix: keep the popout topbar CSS rule intact so the popout test still matches

The display-mode selector was folded into the popout/fullscreen topbar
rule, which broke the exact-string check in stage_tracker_popout.test.js.
Give display mode its own rule instead.

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

* πŸ› fix: keep the popout section-hiding CSS rule intact for the popout test

Same cause as the topbar rule: folding display-mode into the shared
popout/fullscreen selector broke the exact-string check in
stage_tracker_popout.test.js. Give display mode its own rule.

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>
* πŸ› fix: match /.., /./ and ~/. spellings in destructiveCmdRe

The root alternative only accepted a single trailing dot and the home alternatives none, so equivalent spellings of / and $HOME bypassed the check. Widen both to repeated slash/./.. components and make the comment describe exactly what is matched.

Closes #15917
Closes #15922

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

* changelog: fragment for #15952

Generated by changelog-fragment-guard from the PR title (#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>

* πŸ› fix: single bulleted changelog fragment; match /./* and /../* too

Review fixes for #15952:
- fold the two changelog fragments into one `fixed-15952.md` whose first
  line is a `- ` bullet ending in (#15952), as changelog-fragment-guard
  requires; drop the slug-named prose fragment
- append `\*?` to the root alternative of destructiveCmdRe so
  `rm -rf /./*` and `rm -rf /../*` are flagged, mirroring the home branch

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>
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>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* ✨ feat: announce RFC documents proposed as pull requests

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

* πŸ§ͺ test: use keyed Event literals in TestFormat

Event gained Thread and LinkedRFC fields; the positional literals in
TestFormat no longer typecheck under golangci-lint.

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

* πŸ“– docs: mention RFC document PR notices in rfc_channel_id reference

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>
…on authority (#15958)

Signed-off-by: Andy Anderson <andy@clubanderson.com>
Automated pin bump (#8419). Digests recomputed from the downloaded
release artifacts by src/scripts/cli-pin-bump.sh:

  VERSION=2.1.316

Signed-off-by: hive-release[bot] <hive-release@hive.kubestellar.io>
Signed-off-by: clubanderson <407614+clubanderson@users.noreply.github.com>
Co-authored-by: hive-release[bot] <hive-release@hive.kubestellar.io>
…er (#15962)

* πŸ› fix: flag recursive rm of root regardless of option spelling or order

destructiveCmdRe only recognised rm when every option before the target
was a short cluster from [rRfF], so --no-preserve-root, --, long options
and extra flags such as -v let a root or home wipe through to the host
text scanner alone. An argv-based check next to the regex reads the
options the way rm does; the regex keeps covering mkfs, dd and the fork
bomb, and the violation message is unchanged.

Closes #15961

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

* fix: share root target pattern between rm regex and argv check

The v5 merge brought in #15952, which widened destructiveCmdRe's root/home
target set (/.., /./, /../*, ~/., $HOME/.), but rmRootTargetRe used by
isDestructiveRm kept the old set. Spellings only the argv path understands
(--, --no-preserve-root before the path, long options) combined with the
new targets were flagged by neither detector.

Build both regexes from one rmRootTargetPattern constant so they cannot
drift, and add the bypass spellings to the must-flag test list.

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>

* fix: match rm by base name so /bin/rm -rf / is flagged too

isDestructiveRm compared each field to the literal "rm", so an rm named
by path (/bin/rm, /usr/bin/rm) was never read as an rm invocation. The
regex only covers the "-rf <target>" spelling, so /bin/rm -rf
--no-preserve-root / and /bin/rm -rf -- / reached the host text scanner
as the only defence. Match the command word by its base name (quotes
stripped) and cover both the flagged and the clean path forms in tests.

Hive-Fix-Attempt: 2/6
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>
…#15967)

* ✨ feat: bounded scrub-drop Matrix line source behind dashboard.matrix

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

* 🌱 chore: allow pkg/dashboard to import crux/logscrub for the Matrix line source

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>
…15969)

* ✨ feat: display-mode overlay controls and URL-defined view rotation

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

* πŸ› fix: playlist tick via dashboardSetInterval; re-serialise Copy link views

Review fixes for #15969:

- stageTrackerPlaylistInit registers its 1 s tick with the shared
  dashboardSetInterval registry instead of a bare setInterval, keeping
  static/index.html under the setInterval ratchet
  (TestNavbarUpgradeBeeDoesNotAddBareSetInterval). The index is
  wall-clock maths, so pausing while hidden changes nothing.
- stageTrackerDisplayLinkUrl takes the parsed playlist and rebuilds each
  am.view from { filter, group } through stageTrackerViewUrl, so nested
  tokens, unknown keys and rejected views never ride along; interval is
  re-serialised from intervalMs. Uses STAGE_TRACKER_PLAYLIST_PARAM,
  STAGE_TRACKER_PLAYLIST_INTERVAL_PARAM and STAGE_TRACKER_URL_PREFIX.
- Overlay titles live in STAGE_TRACKER_DISPLAY_CONTROL_LABELS (reusing
  STAGE_TRACKER_FULLSCREEN_LABELS.exit.title); the fullscreen / wake-lock
  titles are no longer duplicated in STAGE_TRACKER_DISPLAY_CONTROLS_HTML.
- The Esc/d hotkey leaves through stageTrackerDisplayExit like the
  overlay control.
- Tests: controlsSrc includes the playlist helpers block and covers
  nested-token stripping; the playlist harness rejects a bare
  setInterval; the two-view rotation test no longer depends on which
  view the wall clock lands on at init.

Regenerated static/index.html.

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

---------

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Signed-off-by: ci-maintainer <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>
Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
…s why a human is wanted (#15871)

* fix: stage tracker shows the hive is answering Let hive decide and says why a human is wanted

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: clubanderson <407614+clubanderson@users.noreply.github.com>

* fix: guard stageTrackerActorTip's stageTrackerHumanWhy call for lifted test fixtures

Seven dashboard test fixtures lift stageTrackerActorTip by name without
stageTrackerHumanWhy, so the shuffle job failed with ReferenceError.
Follow the file's existing guarded-helper pattern and regenerate index.html.

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

* test: lift stageTrackerPRListedOwnRow into the row-actions preamble

stageTrackerLeadActionOnRow now calls it, and the node fixture that lifts
functions by name raised ReferenceError without it.

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

* fix: busy Let hive decide keeps its colour, only its own pend busies it, chip names the lead

Review of #15871:

- The busy lead rule lost the cascade to the equal-specificity
  aria-disabled rule that followed it, so the button rendered faint and
  dashed while the hive was asked. The rule is now the compound
  [aria-busy][aria-disabled] selector and sits after the disabled one.
- stageTrackerPendingSync busied the delegate button for any pending
  entry on the row (a confirmed action's "Adopting…", any β‹― action) and
  its restore dropped a pre-existing aria-disabled, making a lead the row
  rendered disabled clickable. The preview's pend is now tagged with the
  button it busies (stageTrackerSetPending's new busy argument); only a
  lead with that data-ham-act goes busy, and its original aria-disabled
  and title are saved and put back when the pend clears.
- The passive chip fires for every PR-targeted lead, so its tooltip now
  names the lead it stands in for (Merge now, Adopt…) instead of always
  saying Let hive decide.
- Tests: node cases cover the covering-PR β†’ passive path (including a
  Merge now lead), and busy/restore under stageTrackerPendingSync (tagged
  vs untagged pend, other verb left alone, disabled lead restored with
  its reason); the CSS test asserts the compound rule follows the
  disabled one. The DOM stub gains removeAttribute, dataset and the
  row-head descendant selector.

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

* fix: covered-issue chip only when the PR is a drawn row

stageTrackerPRListedOwnRow read _stageTracker.items, but a PR nested_under
a listed issue (stage_tracker_links.go) is folded into that issue's sub-row
by stageTrackerVisible and never drawn as its own row outside the Pull
requests view β€” so the issue row showed a passive "decision pending on #N"
chip pointing at a row that did not exist and lost its only decision
button. The render now records the drawn rows (_stageTracker.rows) and the
chip follows them; without a render the items stand in, minus PRs nested
under a listed issue. Test covers nested_under and the drawn-rows path.

Hive-Fix-Attempt: 2/6
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: ci-maintainer <280983584+hivecommons-hive[bot]@users.noreply.github.com>

* πŸ› fix: one preview caption, rows-only chip rule, operator doc for #15773

Review follow-ups on #15871:

- The Let hive decide preview pend now carries its own entry in
  STAGE_TRACKER_ACTING_CAPTIONS ('decide:delegate:preview' β†’
  "Asking the hive…"); the busy lead button renders the pend's caption,
  so the row caption and the button say one phrase and the read-only
  preview is told apart from the confirmed "Hive deciding…". The
  STAGE_TRACKER_LEAD_BUSY_TEXT constant is gone.
- stageTrackerPRListedOwnRow reads only _stageTracker.rows (what the
  last render drew); the harness-only items fallback that duplicated
  stageTrackerVisible's nesting rule returns false instead, and the node
  test sets rows as stageTrackerRender would.
- activity-monitor.md describes the "decision pending on #N" chip, the
  pre-preview "Asking the hive…" busy button, and the human-move tooltip
  fallback to human_merge_reason_text / reason.

Hive-Fix-Attempt: 3/6
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: ci-maintainer <280983584+hivecommons-hive[bot]@users.noreply.github.com>

* πŸ› fix: 085 stage_actor.why says why; drawn-page rows ride ctx, not the payload

stageTrackerActorTip shows stage_actor.why whenever the server sets it, and
the server set a bare "a human decision is owed" on every 085 row, so the
stageTrackerHumanWhy fallback never ran and the πŸ‘€ marker stayed
unexplained. The NeedsDecision / IssueNeedsDecision arm now appends the
lane reason in plain words (as the caption does), else the decision's own
question β€” covered by a direct unit test on stageTrackerStageActor.

The rows the render draws are no longer stashed on the server payload
object: tracker-filter-render.js keeps them in _stageTrackerRows (the drawn
page, group headers aside, so "decision pending on #N" never names a row on
another page), stageTrackerLeadButtonHtml hands them over as ctx.rows, and
stageTrackerPRListedOwnRow(issue, t, rows) is genuinely pure. The passive
chip drops the unused ham-decision-pending class.

Refs #15773

Hive-Fix-Attempt: 4/6
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: ci-maintainer <280983584+hivecommons-hive[bot]@users.noreply.github.com>

* πŸ› fix: covered issue chips to the row that draws its PR; row hover says the decision reason once

- stageTrackerLeadActionOnRow: a secondary covered issue (hive#15503 under
  #15502) whose PR is drawn_under another listed issue renders the passive
  "decision pending on #N" chip instead of a second Let hive decide button;
  the chip names the row that draws the PR. Fixture added to
  TestStageTrackerCoveredIssueDropsPRChipNode.
- stageTrackerTooltip: when the 035 next-move why repeats the caption, the
  Next move line keeps only who moves; the chevron hover keeps the whole why.
- stageTrackerLeadBusySync stashes label/title/disabled in data-ham-*
  attributes like the rest of tracker-*.js; the node DOM stub drops dataset.
- activity-monitor.md documents the server-side 035 why (lane reason, else
  decision_needed.question) and reads "Hive deciding…" like the code.

Hive-Fix-Attempt: 5/6
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: ci-maintainer <280983584+hivecommons-hive[bot]@users.noreply.github.com>

---------

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Signed-off-by: ci-maintainer <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: clubanderson <407614+clubanderson@users.noreply.github.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
…reporting them unclassified (#15988)

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
* fix(dashboard): declare --sp-* spacing ramp on :root

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

* test(dashboard): stop pinning --sp-7 inside .config-modal

The --sp-* spacing ramp now lives on :root (TestSpacingRampDeclaredOnRoot),
so TestAppearanceThemeHoverCannotResizeSettingsModal must no longer require
it inside the .config-modal block.

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

* test(dashboard): re-scope #15920 to a tokens.css regression test

tokens.css already declares --sp-0..9 on :root and head.html links it
before the inline stylesheet, so the ramp was never modal-only. Restore
the .config-modal pin (and its --sp-7 assertion) that keeps a hovered
theme preview from resizing the settings modal, drop the duplicate
:root ramp from base.css, and replace the new test with one that pins
the real invariant: tokens.css supplies the ramp globally, it loads
first, and .config-modal keeps its local pin.

Hive-Fix-Attempt: 2/6
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: ci-maintainer <280983584+hivecommons-hive[bot]@users.noreply.github.com>

* test: drop duplicate .config-modal --sp-* pin check from tokens root test

TestAppearanceThemeHoverCannotResizeSettingsModal already asserts the modal
pin with a guarded terminator lookup; the copy here used an unguarded
strings.Index slice that would panic if the terminator comment were renamed.

Hive-Fix-Attempt: 3/6
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>
Signed-off-by: ci-maintainer <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>
…ft and reduced motion (#15970)

* 🌱 chore: share one chrome-less layout class in popout.css

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

* ✨ feat: display-mode stale dimming, density=tv, burn-in shift and reduced motion

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

* fix(dashboard): route the display-mode clock through formatDashboardTime

TestDashboardDatesUseSharedTimezoneHelpers bans direct toLocaleTimeString
calls in static/index.html so the display clock honours the dashboard
timezone preference; use the shared helper and regenerate index.html.

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

* test(dashboard): load stageTrackerSetFullscreenLayout into the display-controls harness

toggleStageTrackerFullscreen now calls the shared layout helper; the
controls harness in stage_tracker_display.test.js evaluated the toggle
without it and failed with ReferenceError.

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>
* fix: resolve sentinel repo privacy from the repository record, not the PR payload

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

* changelog: fragment for #15980

Generated by changelog-fragment-guard from the PR title (#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 redundant unbulleted fragment for #15980

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>
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>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
…#15993)

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
…ource on Server.Close (#15982)

* πŸ› fix: drop key=value secrets from Matrix lines and stop the matrix source on Server.Close

matrixAcceptLine let short secrets like 'password: hunter2' through when the path was not denied; add a key=value drop rule that leaves Go ':=' and plain prose alone. Server.Close now also closes the matrix source so its run goroutine does not leak (close is already idempotent).

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

* πŸ› fix: match quoted keys in the Matrix key=value secret rule

matrixKeyValueSecret required the key to be followed directly by the separator, so JSON, PHP and Ruby forms like "password": "hunter2" or 'password' => 'x' slipped through. Allow an optional closing quote after the key and cover those shapes in TestMatrixAcceptLine.

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>
…15992)

* test: share the expected --sp-* ramp between dashboard token tests

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

* 🌱 test: reference spRamp[7] from the settings-modal pin test

TestAppearanceThemeHoverCannotResizeSettingsModal still hard-coded
"--sp-7: 20px;" even though spRamp was introduced so both tests share
the ramp. Point it at spRamp[7] so the doc comment is accurate and a
ramp change only needs one edit.

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>
…cide (#15991)

* πŸ› fix: skeleton preview, plan cache and human reasons for Let hive decide

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: clubanderson <407614+clubanderson@users.noreply.github.com>

* πŸ› fix: keep issue numbers out of the tracker-cues CSS block

TestStageTrackerCuesWiring scans the tracker-cues CSS for raw colours by
substring ("#1" etc.), so the two comments this branch added that cite
hive#15773 and "Β· #15759" tripped it in the shuffled dashboard run.
Reword the comments and regenerate index.html.

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

* 🌱 test: feed the 090 human_reason check its lane reason, not the output code

prstage.HumanMergeReason only maps contributor-lane:workflow-path to
the workflow-files code; passing the code itself fell through to
policy, so the actor said "needs a person to merge" and the test
failed on every run.

Hive-Fix-Attempt: 1/6

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

---------

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Signed-off-by: ci-maintainer <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>
Co-authored-by: clubanderson <407614+clubanderson@users.noreply.github.com>
…5995)

* docs: describe what dashboard.matrix feeds in the operator reference

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

* docs: clarify dashboard.matrix buffer has no reader yet

Hive-Fix-Attempt: 1/6

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

---------

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Signed-off-by: guide <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>
hivecommons-hive Bot and others added 6 commits November 3, 2026 15:41
…ument Actions comment deduplication (#16385)

* ✨ feat(notifications): Actions refs sweep skips when the hive:status card carries the condition; docs list deferring workflows

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

* docs(notifications): mark Issue Closed Hint dedup as pending the workflow change

The issue-closed-reopen-hint.yml edit cannot be pushed by the App token
and only exists as a diff on #16363, so the table now says the row is
pending rather than live.

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>

* fix(notifications): match hive:status section keys exactly in hive-status-card.sh

`has` used to test for `<!-- hive:status:section=<arg>` with no
terminator, so `issue-partial:163` matched a card that only carried
`issue-partial:1638` and the refs sweep silently skipped its one-per-issue
nudge for the wrong PR. Parse the rendered keys
(`<!-- hive:status:section=KEY -->`) and require an exact match; an
argument ending in `:` keeps prefix semantics (`issue-partial:`).

Add positive and negative test cases to test-comment-merged-refs.sh and
re-pin TRUSTED_SCRIPTS_TREE for the src/scripts change.

Hive-Fix-Attempt: 2/6
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>

* fix(notifications): hive-status-card.sh filters by author and reuses fetched comments

- `has` only trusts a hive:status card authored by a Bot account, so a human
  quoting the card's markers cannot suppress the refs-sweep nudge.
- `upsert` only edits a comment by HIVE_COMMENT_AUTHOR (default
  github-actions[bot]), never a third party's comment containing the marker.
- comment-merged-refs.sh passes its already-fetched comments via
  HIVE_COMMENTS_FILE, removing the second paginated GET per candidate issue.
- Tests cover the shared fetch, a human-quoted card, and upsert PATCH/POST
  with the author filter and HIVE_COMMENT_AUTHOR override.
- Narrow the changelog fragment to the refs-sweep behaviour this PR ships;
  document the author filters and HIVE_COMMENTS_FILE in notifications.md.
- Re-pin TRUSTED_SCRIPTS_TREE for the src/scripts change.

Hive-Fix-Attempt: 3/6
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>

* chore(governance): re-pin TRUSTED_SCRIPTS_TREE for src/scripts changes

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>
* πŸ› fix(automerge): gate normal-mode stage-note comments

Route stage notes through the transport comment primitive so normal-mode notifications use the canary gate and effects journal. Cover forwarding, write refusal, and missing primitive behavior.

Hive-Run: #16393
Hive-Plan: normal-mode-stage-note-write-gate
Hive-Spec: #16393#normal-mode-stage-note
Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>

* πŸ› fix(automerge): share stage-note comment adapter

Reuse the existing status comment creation adapter for normal-mode stage notes, keeping the single transport write site and fail-closed behavior. Preserve the comment primitive in the superseded-closer test spy.

Hive-Run: #16393
Hive-Plan: normal-mode-stage-note-write-gate
Hive-Spec: #16393#normal-mode-stage-note
Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>

* πŸ› fix(automerge): align notification inventory with shared adapter

Remove the stale direct-producer entry after normal-mode stage notes moved to engineStatusForge.CreateComment, whose existing status classification covers both modes.

Hive-Run: #16393
Hive-Plan: normal-mode-stage-note-write-gate
Hive-Spec: #16393#normal-mode-stage-note
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>
…-only exception (#16400)

* πŸ› fix(automerge): require verified trusted-author commits for the pin-only exception

The human_merge_pin_only exception waived the human-merge-path hold on the
PR author login alone. That login records who opened the PR; any principal
with write access can push onto the branch afterwards, so a pin-shaped
commit from someone else rode a trusted author's PR past the gate. Every
commit behind the head must now be authored by a trusted_authors login and
carry a GitHub-verified signature, and the list must end at the head. The
commit list is read only once the diff already qualifies.

Refs #16399 (upstream action-SHA reachability check remains)

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

* πŸ› fix(automerge): bind pin-only commit provenance to the committer, not the author

GitHub verifies a signature against the committer's email only, so a
verified commit whose author login is in trusted_authors still proves
nothing about the author: a write-access principal with their own
signing key can author a commit as a bot's noreply address, commit and
sign it as themselves, and GitHub reports author.login=<bot>,
verified=true. The pin-only exception now also requires each commit's
committer login to be web-flow (GitHub created and signed it, so GitHub
set the author) or a trusted_authors login (a human trusted author
signing locally). Dependabot/Renovate commits and locally signed
trusted-author commits still qualify.

Fixtures carry a committer; both the sweep and /approve tests gain a
trusted-author/foreign-committer/verified case that is held, and the
sweep test covers a missing committer and a trusted local committer.
Docs, operator reference, changelog fragment and the PinOnlyException
type comment describe the committer requirement; hive-merge.md cites
the open action-sha anchoring item of #16399.

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>
Preserve both histories while integrating v5 snapshot 6398b61, frozen at 2026-11-03T15:39:53Z. Independently reviewed tree f08efe9 retains v6-specific behavior and combined generated assets. Destination refresh and the final bounded incoming fixes remain separate reviewed merge steps before publication.

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
Preserve ancestry through v5 snapshot 03b781e, frozen at 2026-11-03T16:00:38Z. Include the canonical commit-provenance proof, gated stage notes, and refs notification deduplication. Parent source review verified exact canonical proof bytes, retained v6 approval behavior, and the combined scripts-tree pin. Destination integration remains separate before publication.

Signed-off-by: Andrew Anderson <andy@clubanderson.com>
Retain the destination TIP-notice update and all refreshed v5 ancestry. Final incoming v5 is 03b781e; destination is d2dc0f6. Production and regression deltas reviewed without changing gates or coverage floors. Fresh final-head canonical CI and genuine provider admission remain required.

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

Copy link
Copy Markdown
Contributor
6614361

ℹ️ Sentinel notice β€” informational; author is trusted (owner), merge is not blocked

Hive flagged this PR (author @clubanderson, head 66143614dbed) because it matches behaviors that can override security controls, escalate privileges or damage the codebase. This is a heuristic, not an accusation β€” a maintainer should confirm the change is intended before it merges.

  • sensitive_path β€” changes 48 sensitive path(s) (Touches a sensitive path (OWNERS, workflows, policies, hive config, security docs, …))
    • .github/dependabot.yml
    • .github/release-lines.yml
    • .github/workflows/dashboard-mobile-shots.yml
    • .github/workflows/docker.yml
    • .github/workflows/gh-aw-compile.yml
    • .github/workflows/go-security-analysis.yml
    • .github/workflows/integration-candidate-prepare.yml
    • .github/workflows/integration-candidate.yml
    • .github/workflows/integration-technical-go-security-analysis.yml
    • .github/workflows/integration-technical-v2-ci.yml
    • .github/workflows/integration-technical-v2-tests.yml
    • .github/workflows/monthly-outcomes.yml
    • .github/workflows/native-queue-policy.yml
    • .github/workflows/pr-auto-update-test.yml
    • .github/workflows/pr-auto-update.yml
    • .github/workflows/pr-verifier.yml
    • .github/workflows/v2-ci.yml
    • .github/workflows/v2-tests.yml
    • .github/workflows/weekly-competitor-report.lock.yml
    • .github/workflows/weekly-competitor-report.md
    • Justfile
    • dashboard/package-lock.json
    • src/Dockerfile
    • src/Dockerfile.contributor
    • src/crux/go.mod
    • src/deploy/ci-runners/lke/README.md
    • src/deploy/ci-runners/lke/hive-runners-lke-values.yaml
    • src/deploy/gh-aw/README.md
    • 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/entrypoint.sh
    • src/deploy/stream/fleet/configmap.yaml
    • src/deploy/stream/fleet/deployment.yaml
    • src/deploy/stream/fleet/kustomization.yaml
    • src/deploy/stream/fleet/namespace.yaml
    • src/deploy/stream/fleet/networkpolicy.yaml
    • src/deploy/stream/fleet/secret.yaml
    • src/deploy/stream/k8s/deployment.yaml
    • src/deploy/stream/overlays/hive-hosted-matrix/kustomization.yaml
    • src/deploy/stream/public-safe-gate.sh
    • src/deploy/test_arm64_lane_probes_pr_code.sh
    • src/deploy/test_ci_build_lane_aggregate.sh
    • src/deploy/test_pr_verifier_advisory.sh
    • src/go.mod
    • src/proxy/package-lock.json
    • src/proxy/package.json
  • permission_escalation β€” widens workflow permissions or uses pull_request_target (Widens GitHub Actions permissions or switches to pull_request_target)
    • .github/workflows/integration-candidate-prepare.yml
    • .github/workflows/native-queue-policy.yml
  • ci_gate_weakening β€” CI configuration is deleted, softened or has gate checks removed (Disables or softens CI gates (continue-on-error, || true, removed checks))
    • .github/workflows/go-security-analysis.yml
    • .github/workflows/integration-technical-go-security-analysis.yml
    • .github/workflows/integration-technical-v2-ci.yml
    • .github/workflows/integration-technical-v2-tests.yml
    • .github/workflows/pr-auto-update-test.yml
    • .github/workflows/v2-ci.yml
    • .github/workflows/v2-tests.yml
    • .github/workflows/weekly-competitor-report.md
  • security_policy_edit β€” edits security policy, scanner config, rulesets or hooks (Edits SECURITY.md, dependabot/CodeQL config, rulesets or git hooks)
    • .github/dependabot.yml
  • remote_code_execution β€” 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/deploy/test_ci_build_lane_aggregate.sh
    • src/pkg/dashboard/matrix_source_test.go
    • src/pkg/github/automerge/automerge_human_paths_test.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.

@hivecommons-hive

Copy link
Copy Markdown
Contributor

CI triage β€” head 6614361 (quality; read-only, honouring the owner's no-push direction above)

v6 HEAD is green, so everything red here is in the v5β†’v6 union. Besides the already-triaged DCO/provider verdict (and native-queue-policy, which only re-reports it), this head introduces four new, code-level failures that will not clear on their own:

  1. test (rest 3/4) β†’ pkg/github (job)
    • TestEnrichCIStatus_RequiredRedHonoursTheRequiredSet β€” required_check_red_test.go:69: HasFailingRequiredCheck = false, want true (failing [Mutation check on changed lines (advisory)], required map[validate:true]) and :73 RequiredChecksAllReported = true, want false. The v6 test and the incoming v5 required-set semantics disagree on how a non-required failing check is reported.
    • TestGraphQLCheckRunsCarryDatabaseID β€” external_ci_deferral_test.go:173: status context id = 0, want positional 2 (a status has no run id). v5's GraphQL check-run id plumbing landed without the positional-id fallback the v6 test asserts.
  2. test (rest 4/4) β†’ crux/agent/panesignal (job) β€” TestBlockingPromptKey_KnownPrompts β€” blockingprompt_test.go:127: prompt not recognised: a v6-only prompt in the test table is missing from the merged known-prompts list.
  3. NOTICE matches the module graph (job) β€” golang.org/x/time v0.15.0 β†’ v0.16.0 bumped in src/go.mod but NOTICE was not regenerated. Mechanical: src/scripts/generate-notice.sh NOTICE.
  4. relative links and anchors resolve (job) β€” src/docs/api-reference.md line citations drifted against pkg/dashboard/server.go (lines 1794–1814, /api/health … /api/overview/prs.json). Mechanical: bash src/scripts/check-api-reference-citations.sh --fix.

coverage is downstream only (Coverage requires successful test shards (... rest=failure)). Items 3–4 are one-command regenerations; items 1–2 need a union-resolution commit on this branch. Not pushing anything.


🐝 Hive Agent: quality | Instance: hosted-available-oke-11-placeholder-r05x | SHA: unknown

β€” hive: agent=quality backend=copilot model=claude-fable-5.1 copilot=1.0.88

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Checked out head 66143614 and traced the four red checks. Three are mechanical; one is a real regression the merge resolution introduced.

1. omp splash detection regressed (test (rest 4/4) β†’ crux/agent/panesignal FAIL) β€” medium

  • src/crux/agent/panesignal/blockingprompt.go:131-135 is the v5 matcher from #16356: it requires both "welcome to omp" and "press enter to skip".
  • v6 carried a different matcher from #11537, ompSplashSkipRe = (?i)press (?:⏎|enter) to skip at origin/v6:src/crux/agent/manager_pane_signals.go:476, used at :299-306. That regex exists nowhere in the merged tree (grep -rn ompSplashSkipRe src/crux is empty).
  • The merged test keeps v6's cases at blockingprompt_test.go:117-118 ("press ⏎ to skip" and "press enter to skip" β†’ "Enter"), and both fail with prompt not recognised at :127, exactly as CI shows (TestBlockingPromptKey_KnownPrompts/omp_splash_unicode_key, /omp_splash_plain_key).

Consequence: a v6 hub running omp with its default unicode symbol preset renders press ⏎ to skip; the merged matcher never matches it, so the splash is no longer auto-skipped and the pane wedges at startup β€” the behaviour #11537 fixed on v6 is lost. The v5 matcher also wants "welcome to omp", which the v6 test fixtures (and the #11537 comment describing the real splash) do not contain. Suggested resolution: make the omp entry's match use the #11537 regex semantics (press (?:⏎|enter) to skip, case-insensitive) so both the v6 fixtures and v5's welcome to omp … press enter to skip text pass. This should be fixed in the merge commit rather than deferred, since it is a v6 feature silently reverted.

2. NOTICE matches the module graph β€” one hunk: NOTICE:6408-6411 still records golang.org/x/time/rate v0.15.0 while src/go.mod now has golang.org/x/time v0.16.0. src/scripts/generate-notice.sh and commit verbatim.

3. relative links and anchors resolve β€” reproduced locally: bash src/scripts/check-api-reference-citations.sh β†’ 32 drifted citation(s) (merged pkg/dashboard/server.go shifted the lines cited by docs/api-reference/*.md). --fix rewrites them.

4. DCO / native-queue-policy β€” the probot summary says the GraphQL fallback for a >250-commit PR did not complete; native-queue-policy fails only with Unhandled error: Error: DCO: failure. That is the provider problem the body already routes to #16404, not a missing sign-off I could identify.

I did not review the 1,518-file union beyond these checks. Agree with the body: merge commit only, never squash.

β€” hive: agent=reviewer-2 backend=copilot model=claude-fable-5.1 copilot=1.0.88

Restore the existing v6 OMP splash-footer behavior in the extracted pane-signal leaf. Correct legacy-status fixtures to retain real node/head evidence without inventing check-run identities, and cover required-red, advisory-red and unavailable-status cases without changing production admission policy. Refresh drifted citations and import the byte-bound NOTICE generated by canonical CI for 6614361. Preserve true merge ancestry, exact scripts pin, all coverage floors and external DCO requirements. Fresh remote CI remains required.

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

Copy link
Copy Markdown
Contributor

needs-human: per the owner's direction above (no pushes; DCO App verdict is UNKNOWN at >250 commits and native-queue-policy only re-reports it), agents stand down on this PR until the App can evaluate the full commit list or an operator exception is established.


🐝 Hive Agent: scanner | Instance: hosted-available-oke-11-placeholder-r05x | SHA: unknown

β€” hive: agent=scanner backend=copilot model=claude-fable-5.1 copilot=1.0.88

@hivecommons-hive

Copy link
Copy Markdown
Contributor

Needs a decision, not a merge: the automerge lane will not merge this unattended.

Head: verdict approved @670776c, 82/84 checks green, 2 pending, DCO yes, soak 0s of 10m0s.

How to release it:

  • Answer the needs-human ask β€” the hive note that added the label says what it needs (a supersession decision, a workflow approval, …) β€” then have a maintainer remove that label; the hive takes its own needs-human off when the cause clears, and /hive approve does not remove it.

Clear the cause and the hive re-evaluates on its next pass; the stage moves to stage/090-ready-for-human-merge, stage/095-pre-merge-soak or the merge queue on its own. Approval does not bypass CI, hold labels or other merge gates.

…the top-up

Plain merge of origin/v6 (5c963b7) into the v5->v6 top-up; the only
conflicts were the additive export/import lists in
.github/scripts/native-queue-policy.cjs and its test, resolved to the
union, which is byte-identical to v5 after #16442 (ed048ec).

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

Copy link
Copy Markdown
Member Author

Status from the maintainer side (delegate for @clubanderson), head now 9aa3c149e1a9361427c05059eab9f994e0b0692d:

What changed on this branch. One plain merge commit, 9aa3c149 = merge of origin/v6 (5c963b70, which includes #16467) into the previous head 670776c9. No squash/rebase/force-push/amend; first-parent chain is unchanged (670776c9 ← 66143614 ← 63cfbb0b ← a4df37af …). This was required, not cosmetic: #16467 landed the policy-script half of #16442 on v6 and this PR touches the same module.exports/import lists, so without the merge GitHub would have flagged the PR as conflicting. The two additive conflicts were resolved to the union, which is byte-identical to v5 after #16442 (ed048ecb). The 14 v5 commits that landed after this snapshot are not pulled in here (a full git merge origin/v5 conflicts in 15 Go/JS/docs files and belongs to the next deliberate top-up).

What the admission path does now. For a v6-base PR, native-queue-policy runs the yml from the default branch (v5, now with #16442) and checks out v6's scripts (now with #16467). When the DCO App reports its "complete commit list could not be retrieved … verdict is unknown" diagnostic, the job reports DCO provider unavailable / verdict UNKNOWN with a job summary and still fails β€” exactly what #16404's owner decision asked for. It does not, and must not, admit this PR.

What still blocks this PR (unchanged, external): App 1861 cannot enumerate 307 commits β€” check run 126214935396 on 9aa3c149 fails identically to the five earlier heads (1cb3026f, 96ba258f, 1989d20a, 66143614, 670776c9). The check-run re-request endpoint returns 404 for a user token (same as the owner observed). Until the provider's GraphQL fallback is repaired (dcoapp/app#339 is still unmerged) or an explicitly authorised operator exception exists, admission stays blocked. The needs-human label from the scanner is accurate and is left in place; note that while it is present native-queue-policy fails early on the label ("head/base moved, draft, closed, or held") rather than reaching the UNKNOWN report.

Nothing further for agents to do here; the next action is the owner's (provider recovery or an explicit exception).

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Decision needed: @clubanderson should decide whether to admit this v5β†’v6 forward-merge (merge commit, never squash) while the probot DCO verdict on the head is UNKNOWN per #16404, given no other check is red on 9aa3c14.

Routing to a human; nothing defective found in what I could verify. A 1,522-file forward-merge is not something I can read end to end, so this is a bounded check of the merge mechanics and of the delta the PR body does not describe.

Verified at head 9aa3c149:

  • Ancestry holds. Compare API: v5 snapshot 03b781e3 β†’ head is ahead 1283, behind 0; v6 tip 5c963b70 β†’ head is ahead 307, behind 0. Both are ancestors, so a merge commit lands without rewriting either line.
  • The body is one push stale. It cites head 670776c9; the actual head is 9aa3c149, which is 670776c9 + a plain merge of v6 tip 5c963b70 (parents 670776c9, 5c963b70), bringing in #16437, #16450, #16462, #16467. Worth refreshing the "published head" sentence so the admission record matches.
  • The merge resolution is what the commit message says. .github/scripts/native-queue-policy.cjs at head has blob 51bcf634…, identical to v5 ed048ecb (#16442). The v6-side hunks all survive the merge: ProviderUnavailable class and the dcoEnumerationUnavailable(run) throw in native-queue-policy.cjs; standbyLaneItems(...) in src/pkg/dashboard/contribute_ws.go and its admissionQueueSnapshot(readyQueueDefaultLimit, withheldNone).queue caller in src/pkg/dashboard/server.go; the standbyPush guard in bin/contributor-relay.js. diagnose-dco-provider.cjs, standby_qualified_test.go, v6-readiness-runbook.md and dco-provider-unavailable.md are byte-identical to v6 tip.
  • Checks. 76 pass on this head. The two red ones are probot DCO (UNKNOWN β€” the >250-commit enumeration gap tracked in #16404) and native-queue-policy, which failed with #16378: head/base moved, draft, closed, or held (native-queue-policy.cjs:83) β€” a stale-run eligibility abort, not a policy verdict against this head.

I have not reviewed the 1,500+ imported v5 files individually; they landed on v5 through their own reviews. The one thing only a maintainer can settle is the DCO-provider question above. I have not approved or closed anything.

β€” hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88

@clubanderson

Copy link
Copy Markdown
Member Author

Operator exception β€” manual merge-commit merge of #16378 into v6

Authorized by: @clubanderson (owner), verbatim: "Operator exception: I authorize a manual merge-commit merge of #16378 into v6 now (no squash/rebase; documented in a comment)" β€” executed by his delegate.

Why an exception is needed. The only non-green admission evidence on this PR is the external DCO App (App 1861): with 307 commits the REST list stops at 250 and the App's GraphQL fallback does not complete, so it reports "the DCO verdict is unknown" β€” a provider-unavailable condition, not a missing sign-off. It has done so identically on six consecutive heads (1cb3026f, 96ba258f, 1989d20a, 66143614, 670776c9, 9aa3c149; latest check run 126214935396); the check-run re-request endpoint returns 404 for user tokens; the upstream repair (dcoapp/app#339) is unmerged. The prow dco status on the same head is success ("All commits have Signed-off-by"). native-queue-policy only re-reports the provider result (and, with needs-human present, stops at the label). #16442 (v5, ed048ecb) and #16467 (v6, 5c963b70) were landed first so this condition is reported honestly as UNKNOWN rather than as a sign-off failure; neither relaxes admission, which is why an explicit owner exception is the only sanctioned path.

Verified on exact head 9aa3c149e1a9361427c05059eab9f994e0b0692d before merging:

How it is merged: GitHub merge commit (gh pr merge --merge). No squash, rebase, force-push or amend at any point in this PR's life; the branch's first-parent chain (9aa3c149 ← 670776c9 ← 66143614 ← 63cfbb0b ← a4df37af …) is preserved and becomes the second parent of the new v6 merge commit, so every v5 commit keeps its original SHA and Signed-off-by.

Scope: this exception applies to #16378 only. #16404 stays open for the upstream enumeration bug; the admission gate and App permissions are unchanged. Removing needs-human (scanner, 20:48Z) since its ask β€” "until the App can evaluate the full commit list or an operator exception is established" β€” is now answered.

@clubanderson
clubanderson merged commit baf3e1f into v6 Nov 3, 2026
98 of 101 checks passed
@kubestellar-prow
kubestellar-prow Bot deleted the sync/v5-to-v6-20261102 branch November 3, 2026 22:54
@github-actions

github-actions Bot commented Nov 3, 2026

Copy link
Copy Markdown
Contributor

Thank you for your contribution! Your PR has been merged.

We'd love to hear how your experience was: share feedback

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

conflicts-with-open-pr Open PR has a merge conflict with another open PR dco-signoff: yes Indicates the PR's author has signed the DCO. size/XXL Denotes a PR that changes 1000+ lines, ignoring generated files.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants