Repository navigation
π± chore: v5 β v6 top-up (2026-11-03) - #16524
Conversation
β¦pin-only exception admits them (#16435) * π fix(automerge): verify action-sha pins resolve upstream before the pin-only exception admits them ClassifyPinOnlyActions returns the accepted action-sha pins; the pin-only exception resolves each against its upstream repo (tag ref with annotated-tag dereference, or ancestry of the default branch) and holds on any failure. The pin-only and DCO commit listings use the exported pkg/github.ListPRRepositoryCommits helper. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com> * π§ͺ test(intent): separate pin-replacement pairs with context in accepted-pins fixture The fixture placed two -/+ pairs back-to-back, which the bounded pin grammar correctly refuses as an interleaved replacement. Insert a context line between the pairs and widen the hunk header to 5/5 so the fixture exercises the dedup path it was written for. 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>
β¦ce, skip unchanged edit (#16445) * π fix(advisory): re-pin digest past foreign-authored comment, warn once, skip unchanged edit When a human-authored digest comment sits newer than the App's own on the advisory issue, the scan warned on every ~5 min cycle and never settled. Resolve by never selecting a non-App comment, re-pinning to the App's own digest (or creating one) and persisting the anchor id, logging the foreign comment once per issue per process (Debug thereafter) with the action taken, and skipping the PATCH when the material fingerprint for the anchor is unchanged. Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π docs: refresh client.go citation range in scale-envelope.md Signed-off-by: Andrew Anderson <andy@clubanderson.com> --------- Signed-off-by: Andrew Anderson <andy@clubanderson.com> Co-authored-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
β¦nd report lost prompts as task_failed (#16447) 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: onkar717 <144542684+onkar717@users.noreply.github.com>
β¦r the matrix streamer (#16448) * π stream: deeper capture queue, wallclock PTS and a stall watchdog for the matrix streamer The Twitch player buffers while the encoder reports fps=30 speed=1x. Harden the streamer so transient encoder hiccups no longer become PTS gaps and so stalls restart ffmpeg instead of waiting for the liveness probe: - VIDEO_QUEUE_FRAMES default 16 -> 256 (2.1 GiB worst case, inside the 6 Gi limit); AUDIO_QUEUE applies -thread_queue_size to both audio inputs - -fflags +genpts -use_wallclock_as_timestamps 1 on x11grab and -af aresample=async=1 so audio/video PTS cannot drift - watchdog samples the -progress file every 10 s, SIGTERMs ffmpeg after 30 s without frames or 60 s of speed < 0.90x, and writes the last 20 samples, last stall and restart count to /run/stream/health.json; healthcheck.sh surfaces that state - README: ingest-RTT guidance and a 'Why does the player buffer?' section with the diagnosis recipe Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π stream: leave k8s/deployment.yaml untouched The memory-budget comment edit tripped TestHivecommonsStreamConfigUnchanged, which pins the live stream's manifests byte-for-byte. The queue math is documented in entrypoint.sh and the README instead. Signed-off-by: Andrew Anderson <andy@clubanderson.com> --------- Signed-off-by: Andrew Anderson <andy@clubanderson.com> Co-authored-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
β¦uppress reviewer-lane no-findings reviews (#16434) * π Edit claim ledger, needs-decision and sentinel comments in place; suppress reviewer-lane no-findings reviews Issue-side comment churn (#16433): hive:claim/hive:release transitions now maintain one per-issue claim-ledger comment edited in place (github.notifications.claim_ledger, default on) that every existing marker reader still parses; the needs-decision/ready entry comment is rewritten with an Updated line on re-entry; a new push rewrites the hive's existing sentinel comment instead of stamping the PR again; and in quiet mode a reviewer-lane review that narrates "no findings" with an approving, finding-free verdict is recorded and folded into the hive:status comment instead of being submitted. Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π Fix test literal and adoption read counts for the entry-comment upsert Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π Teach the claim hook tests the ledger's comment lookup; cover ledger vs legacy mode Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π Inventory EditIssueMarkerComment as a status passthrough; keep marker openers out of ledger release prose Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π± Cover EditIssueMarkerComment create, edit, no-op and foreign-author paths Signed-off-by: Andrew Anderson <andy@clubanderson.com> --------- Signed-off-by: Andrew Anderson <andy@clubanderson.com> Co-authored-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
β¦ops a safety gate the previous runtime overlay carried (#16449) * π fix(config): warn, report and optionally refuse when a seed boot drops a safety gate the previous runtime overlay carried A spoke booting from the ConfigMap seed (runtime overlay moved aside, truncated or never written) silently lost review.require_labels, project.paused_repos, writeback.per_repo and project.issue_filter.require_labels. The boot now compares the effective config against the previous overlays still on the PVC (hive.yaml.runtime, hive.yaml.runtime.bak, hive.yaml.dashboard), logs one 'WARN safety gate dropped' per gate with the previous value, serves the list on /api/status as config.gate_regressions, shows a Settings banner, and with config.refuse_gate_regression: true refuses to start (exit 21). Fixes #16438 Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π fix(config): rename gate-regression test helpers to avoid clashing with agent_sandbox_gate_test Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π fix(config): pin config: block in agent-runtime golden and move gate banner margin into CSS TestAgentRuntimeConfigGolden saw the new config: block as an unexpected key; TestStyleRatchet counted the banner's inline style= attribute above baseline. Signed-off-by: Andrew Anderson <andy@clubanderson.com> --------- Signed-off-by: Andrew Anderson <andy@clubanderson.com>
Every method in crux/api/lanes_client.go read 0% in the crux/api coverprofile: the lane client is only exercised from crux/serve and pkg/laneparticipant, which the per-package floor does not count. The new test pins the routes, path escaping, request bodies, the 204 nil-claim contract and lane error-code surfacing against a fake server, the same way client_test.go pins the agent routes. Signed-off-by: quality <280983584+hivecommons-hive[bot]@users.noreply.github.com> Co-authored-by: quality <280983584+hivecommons-hive[bot]@users.noreply.github.com> Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
β¦ression test (#16451) Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com> Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
* π Restore activity monitor layout and compact chevrons Restore canonical metadata spans, full-row right-anchored controls and bounded shared tracks for linked and multi-PR rows. Separate current-phase position from running activity. Wrap chevrons to two visible lines with accessible full details and icon-only unknowns. Preserve decision controls without competing tooltip overlays. Independent review cleared the final implementation; canonical CI and published-render geometry/interaction proof remain required. All reported chevron work stays in this single branch and PR. Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π Repair activity monitor regressions and cover dense layouts Exercise production row layouts and decision controls in the existing browser harness, repair coupled expectations without reducing ratchets, and label observed PR source repair Remediating while preserving native state identifiers. Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π Repair activity monitor assertions and browser focus checks Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π Isolate activity browser fixtures and verify captured pixels Signed-off-by: Andrew Anderson <andy@clubanderson.com> --------- Signed-off-by: Andrew Anderson <andy@clubanderson.com>
* β¨ perf: progressively load the dashboard shell and live panels Signed-off-by: Andy Anderson <andy@clubanderson.com> * π fix: initialize deferred dashboard overlays after script setup Signed-off-by: Andy Anderson <andy@clubanderson.com> * π fix: preserve dashboard route and refresh wiring contracts Signed-off-by: Andy Anderson <andy@clubanderson.com> * π fix: coalesce live cost redraws while the dashboard is hidden 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>
β¦16460) The hub stamps hive.hivecommons.dev/* (hiveLabelPrefix in src/pkg/hub/hosted_namespace_identity.go); the runbook still named the legacy hive.kubestellar.io/* keys, so its kubectl selector and jsonpath examples returned nothing. Fixes #16459 Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com> Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Signed-off-by: Andrew Anderson <andy@clubanderson.com> Co-authored-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
β¦ as chevrons (#16464) Restore content-driven chevron height (one line of caption -> one-line chevron, two lines -> two, never more) and centred captions, which #16454 replaced with a fixed two-line grid row and a flush-left stretched copy box. Drop the boxed-card opt-out that #16384 gave the concurrent CI / Review lanes so they share the stage chevron clip-path, padding and height, with lane state expressed through fill and text colour. Geometry probe and CSS-contract tests now fail on the previous behaviour; density fixture gains a one-line held row. Signed-off-by: Andrew Anderson <andy@clubanderson.com>
β¦sion (#16442) Distinguish the authoritative DCO provider's incomplete-history UNKNOWN diagnostic from a sign-off failure while keeping both admission paths closed. Show an operator-directed Actions summary and retain the current App/head/provenance requirements without retries or status substitution. Add a read-only full-history query reproducer, regression fixtures beyond the REST ceiling, and a recovery runbook linking the observed failures and the unmerged provider-side repair candidate. Hive-Run: #16404 Hive-Plan: dco-provider-diagnosis-and-reporting Hive-Spec: #16404#maintainer-fourth-direction Signed-off-by: thatguywob <thatguywob@users.noreply.github.com> Co-authored-by: thatguywob <thatguywob@users.noreply.github.com>
β¦#16468) 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>
Rows are now only as tall as their longest caption (#16464), so the browser geometry probe's requirement that each lane keep the same height across rows contradicted the shipped contract and failed every viewport in the reporting-only screenshot workflow. Within-row uniformity and the content-driven height check remain. Signed-off-by: Andrew Anderson <andy@clubanderson.com>
* π± logscrub: align v5 scrubber categories with v6 Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com> * changelog: fragment for #16470 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: single bulleted fragment for #16470 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>
The CI and Review tracks of tracker column 50 both rendered icon 50 (the eye). Add a flask glyph keyed 'ci' and pick it for the CI track; Review keeps the eye. Regenerate static/index.html and tighten the activity test to assert the two tracks differ. Fixes #16472 Signed-off-by: Andrew Anderson <andy@clubanderson.com>
β¦6480) "CI waiting Β· test not observed" meant nothing to an operator: it was the internal state plus the raw required-check name. Rewrite every CI caption in stage_tracker_activity.go so it says what the state means (which required check, whether it reported, how many are done) and add a Detail field carrying a longer plain-language sentence that the strip appends to the chevron tooltip. Reword the fix-task CI-wait and CI-retry code-activity captions the same way. Fixes #16479 Signed-off-by: Andrew Anderson <andy@clubanderson.com>
β¦6483) An operator's Adopt now on a PR whose author is one of this hive's owners recorded a forced adoption the stage engine refused on every sweep as person-authored (#15202) without ever consuming it, so the row sat at "adoptingβ¦ (queued)" with the action spent. The hive still never adopts a person's own PR on its own initiative, and an adoption/supersede request is still refused for one; an explicit forced Adopt now now bypasses that refusal and proceeds. The attempt record (MarkAdopted) consumes the forced mark as for any other author, so the row reads queued β adopting β adopted. The dashboard audit entry names the author and whether they are a hive owner. Fixes #16482 Signed-off-by: Andrew Anderson <andy@clubanderson.com>
β¦e track the hive waits on (#16488) A column-50 track whose evidence was unknown, stale (observed_at older than 2 min, which server caches routinely are) or for another head rendered as a bare glyph. Both tracks share the row's current-phase tint, so an operator saw two amber chevrons with no words and no way to tell which one was live. stageTrackerActivityTrackHtml now always renders glyph + caption; stale evidence says how old it is, missing evidence says 'no evidence for <sha>'. Freshness still gates the pulse and the green fill. A new pure stageTrackerTrackLead picks the track the hive waits on when both share the phase (review findings / red CI β the fix; otherwise CI until green, then review) and the chevron text carries 'Β· new head <sha>' on the leading CI track or 'Β· waiting on CI|review|the fix' on the other; the tooltip repeats it as a sentence. Fixes #16484 Signed-off-by: Andrew Anderson <andy@clubanderson.com>
β¦, not stuck Actionable (#16486) A 010 issue whose stage reason carries the engine's "waiting on N open child issue(s)" suffix is tracking its children; the scanner leaves it alone. The row said "Actionable Β· waiting for the scanner to pick it up" with an 18h since, which reads as scanner latency. - stageengine.ReasonOpenChildren parses the suffix. - StageTrackerItem.OpenChildren; caption/actor say Tracking β 3 of 4 children done; waiting on 1 open child issue(s); stage_actor is none. - Lit 010 chevron reads Tracking Β· 3/4 (or Tracking Β· N open children). Fixes #16481 Signed-off-by: Andrew Anderson <andy@clubanderson.com> Co-authored-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
β¦16487) * π dashboard: restore compact tracker rows (one-line chevrons, inline title) Tracker rows in the Activity Monitor doubled in height after the layout restore (#16454) and acceptance stabilisation (#16461): - the row toolbar (hold / Let hive decide pills) was a separate full-width grid line between the title and the chevrons, reading as a 30-40px blank band; it now trails the PR cell on the title line; - chevrons carried a two-line caption budget, sp-2 vertical padding and a matching grid-auto-rows floor (24-48px); captions are clamped to one line with ellipsis and sp-1 padding (~20px), the pre-#16454 density; - the issue title had a max-content flex basis inside a wrapping .ham-head-main, so it dropped to its own line whenever it could not fit entirely; it now fills the rest of the chip line (flex: 1 1 0) and wraps only its own words. The PR ref does the same with a 20em floor so pills stay on the first line; - row padding sp-5 -> sp-4. Simple rows measure 59px (was 71-99px) at 1440x1200@1; 16 rows fit in the viewport where 8 did. Tests and the tracker-header-geometry probe are retargeted to the one-line contract; a node test pins the inline title. Fixes #16485 Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π dashboard: retarget Go-embedded tracker row fixture to the inline chip cell TestStageTrackerRows still asserted the pre-PR head markup (a separate <li class="ham-seg cap"> toolbar line); the chips now trail the PR cell (ham-head-pr / ham-head-tools) on the title line. Signed-off-by: Andrew Anderson <andy@clubanderson.com> --------- Signed-off-by: Andrew Anderson <andy@clubanderson.com>
β¦/total, lit while in progress (#16494) * fix(dashboard): CI/Review chevrons carry pass/total and done/total counts, lit while in progress Operator steer on #16488: dull current-phase tint with glyph-only chevrons was unreadable. Every CI/Review chevron in progress or finished on the current head now reads a count caption (CI p/t, Review d/t) with the full active fill, done style when complete, error style on a finding, and the glyph alone only when not started. Server adds checks_passed/total/failed/ failing and perspectives_done/total to StageTrackerActivity. Signed-off-by: Andrew Anderson <andy@clubanderson.com> * test(dashboard): geometry probe expects count words on lit tracks; count captions wrap like peer chevrons The browser probe's 'unknown track is icon-only' assertion predates the count captions; a lit unknown track now reads 'CI ?/? Β· no checks reported yet' and only a not-started (.todo) track is the glyph alone. Drop the nowrap/ellipsis rule so the count caption honours the strip's two-line wrap budget instead of truncating mid-count on phones. Signed-off-by: Andrew Anderson <andy@clubanderson.com> --------- Signed-off-by: Andrew Anderson <andy@clubanderson.com> Co-authored-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
β¦#16490) * fix(inference): log why unidentified translator callers are refused Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com> * changelog: fragment for #16490 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> * fix(inference): move translator refusal attrs to own file; single changelog fragment Keeps github_proxy_inference.go under the crux 800-line ratchet and gives the PR one bulleted changelog entry as changelog-fragment-guard requires. 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>
* π± ci: route FOSSA uploads to the CNCF Hive Commons team Pass project/team/branch to fossas/fossa-action and add a v3 .fossa.yml so the first upload creates hivecommons/hive inside the CNCF FOSSA org "Hive Commons" team (id 207492). FOSSA honors team only on the upload that creates the project, so this must land before FOSSA_API_KEY is set; until that secret exists the workflow keeps skipping. Refs cncf/sandbox#545 Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π± ci: include src/proxy npm module in the FOSSA target set The spoke's runtime Node proxy ships to users, so the first CNCF upload should cover its dependencies; discord/ and .github/scripts/ stay out as repo tooling. Signed-off-by: Andrew Anderson <andy@clubanderson.com> * ci: analyze before test in FOSSA workflow fossa-action v2 with run-tests=true runs only 'fossa test' and forwards team/branch to it, which the CLI rejects (run 41750174253). Always run analyze (the upload that creates the project under the team), then run the policy test as a separate step on pull requests. Signed-off-by: Andrew Anderson <andy@clubanderson.com> --------- Signed-off-by: Andrew Anderson <andy@clubanderson.com> Co-authored-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
The project now exists in the CNCF FOSSA org under the Hive Commons team (locator custom+162/hivecommons/hive), so the badge resolves. Reverts the removal in #13065, pointing at the new locator. Signed-off-by: Andrew Anderson <andy@clubanderson.com> Co-authored-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
#16503) Keep rolling back undelivered claims while logging ErrCloseSent at INFO. Preserve warnings for other send failures and cover both paths with rollback and reassignment regression assertions. Hive-Run: #16496 Hive-Plan: assignment-close-race Hive-Spec: /issues/16496#evidence Signed-off-by: thatguywob <thatguywob@users.noreply.github.com> Co-authored-by: thatguywob <thatguywob@users.noreply.github.com>
Keep one active full collector and fresh-stats collector per heartbeat loop so repeated timeouts cannot accumulate blocked work. Include slow section timings in the original warning captured by fleet self-reports. Hive-Run: #16504 Hive-Plan: heartbeat-collection-containment Hive-Spec: #16504#collect-timeout Signed-off-by: thatguywob <thatguywob@users.noreply.github.com> Co-authored-by: thatguywob <thatguywob@users.noreply.github.com>
β¦glyph, workflow-approval stall caption (#16600) * β¨ Activity Monitor chevrons: active-with-text / icon-only spec, ring glyph, workflow-approval stall Fixes #16599 - 1 <= active <= 2 chevrons per row: the stage column is always lit (at 050 the Review column, even while queued); a CI/Review track lights only while genuinely running on the current head. Inactive chevrons are icon only (neutral / red / green); counts move to the tooltip. - CI/Review icons become a segmented ring (stageTrackerRingSvg) that fills with passed checks / posted perspectives, newest step pulsing while running; tokens only, aria-labelled. - When nothing runs the blocker column is filled in a still stalled style, with a person glyph for human actions: 'CI 1/11 Β· waiting on workflow approval' from pkg/github's new PRCIActivity.WorkflowApproval probe of action_required runs (90s per-head cache). - Review-pending rows never read unknown / default 5/5: stageTrackerReview- PendingState yields reviewing (N/5 in progress) or adjudicating. - Node + Go tests: ring, workflow approval, review pending, row fixtures (#16567 gated, #427 queued, #16456, #16268, both-running) asserting words only on active chevrons and 1 <= lit <= 2. Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π Address #16600 review: bound workflow-approval cache, fix CI - forgetWorkflowApproval now has callers: CI webhook deliveries for a head drop its cached probe; the cache also evicts entries older than 30m so the package-global map stays bounded (reviewer finding). - Detail no longer names a config key that does not exist yet. - Drop trackerReviewPendingStage duplicate of trackerNumberReviewPending. - MotionPreviewSample writes its node script to a temp file instead of node -e (argument list too long on Linux runners). - Fix RankActionableIssues citation drift in scale-envelope.md. Signed-off-by: Andrew Anderson <andy@clubanderson.com> --------- Signed-off-by: Andrew Anderson <andy@clubanderson.com>
The replicas:0 hive-hosted-matrix stream overlay still requested view=matrix, which only works via the deprecated alias in boot.js. Switching to view=reader keeps the overlay working once the alias is removed. Closes #16612 Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com> Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
β¦a restart (#16613) Right after a roll the Overview headline shows the nowrap "refreshing after restart Β· 0/10 repos scanned" pill. Its min-content width stretched the single auto column of .overview-headline, so every KPI tile, subline and sparkline was laid out ~470px wide on a 430px phone and clipped until the scan finished and the pill disappeared. The tiles also read var(--fs-2xs), which was only defined inside .config-modal and fell back to the body size everywhere else. - stale pill: wraps, min-width:0 / max-width:100%, justify-self:start - headline children, kpis grid, tiles, sublines, sparklines: min-width:0, overflow-wrap:anywhere, max-width:100% - define --fs-2xs/--fs-2xl/--fs-3xl at :root with the design-system values - phone (<=720px): tile label/subline use --fs-base / --fs-md Fixes #16601 Signed-off-by: Andrew Anderson <andy@clubanderson.com> Co-authored-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
β¦6598) * β¨ Auto-decide reviewer scope questions on hive-authored green PRs A review perspective that ends with a question rather than a finding parked hive-authored, green, finding-free PRs at 085 on wording alone (docs#618, hive#16456). Classify the ask: policy sign-off stays a person's; a scope / follow-up question runs the existing Let-hive-decide pipeline once per head (approve -> merge queue + follow-up issue; direction -> fixer; owner-only -> park with the recommendation). Config governor.review.auto_decide_questions (default on); review_questions_autodecided_1h on /api/status. Fixes #16597 Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π Resync api-reference citations after server.go shift Signed-off-by: Andrew Anderson <andy@clubanderson.com> * π± Test the patch-key rule on a file outside the policy path list Signed-off-by: Andrew Anderson <andy@clubanderson.com> --------- Signed-off-by: Andrew Anderson <andy@clubanderson.com> Co-authored-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
|
Scanner triage at ad59ff3 (comment-only; not a hive-authored branch): π Hive Agent: β hive: agent=scanner backend=copilot model=claude-fable-5.1 copilot=1.0.88 |
β¦6624) A row flagged needs-human renders its reason in the hive-note line (blocker short + first sentence of the ask + asker + lead option + who is pushing a covering PR), keeps the Blocked Β· <short> π€ word in the issue's decision column when a PR strip drives the row, turns the needs-human pill into a reveal button with the full reason as hover, captions the 085 chevron 'Needs human Β· <short reason>' from a small map of known reasons, reads 'decision on #PR' on an issue blocked only by its PR's pending decision, leads with the first enabled blocker option when the server disables decide:delegate (rendered muted with reason_if_disabled), and drops the double space in the 'owed β blocked: β¦' caption. Fixes #16623 Signed-off-by: Andrew Anderson <andy@clubanderson.com>
Hive-Run: #16611 Hive-Plan: centralize-code-activity-compatibility Hive-Spec: issue-16611#description Signed-off-by: thatguywob <thatguywob@users.noreply.github.com> Co-authored-by: thatguywob <thatguywob@users.noreply.github.com>
β¦16625) * π Keep a not-yet-run current-phase track visibly filled mobile-shots tracker-header-geometry failed on rows whose current-phase CI/Review track has not run yet (unknown / queued beside its running neighbour): the icon-only .todo chevron lost its fill. Give .todo.ham-current the same muted field as .done/.failing β glyph only, neutral β and make the probe's failure message name the cell. Signed-off-by: Andrew Anderson <andy@clubanderson.com> * changelog: fragment for #16625 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: clubanderson <407614+clubanderson@users.noreply.github.com> * π changelog: drop duplicate fragment (fixed-16625.md carries the entry) Signed-off-by: Andrew Anderson <andy@clubanderson.com> --------- Signed-off-by: Andrew Anderson <andy@clubanderson.com> Signed-off-by: clubanderson <407614+clubanderson@users.noreply.github.com> Co-authored-by: clubanderson <407614+clubanderson@users.noreply.github.com> Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
A row stacked its "Needs human: β¦" line over the "Hive: moved to β¦" stage note. Rank every candidate β needs-human / decision > blocker > next move / claim > optional delegation > stage note β and render the winner alone on issue rows and PR sub-rows. A lower note's facts ride the winning line as a trailing clause only when they add something the row does not already say; a stage note that merely restates the lit chevron is dropped. The single line keeps its ⬑ icon, View link and one age. Fixes #16627 Signed-off-by: Andrew Anderson <andy@clubanderson.com>
β¦up to 5e0e398) Conflicts resolved (v5 wins unless v6 intentionally diverged): - src/pkg/config/config.go: v6 split the file; v5 GovernorConfig.Review + GovernorReviewConfig ported to governor_config.go - src/pkg/config/dashboard_matrix.go: v5 comment wording kept, v6 DashboardMatrixConfig name kept (v6 already has a notifications MatrixConfig); dashboard_config.go comment updated to match - src/pkg/dashboard/server.go: v6 RUnlock-before-build kept, v5 ReviewQuestionsAutodecided1h added to /api/status payload - src/pkg/github/issue_unpark_command.go: v6 owner-step menu kept, v5 held-after-decision (#16591) gate added with v6 ensureUnparkNoticeComment signature - src/pkg/scheduler/kickmessage.go: both hints kept (v6 upstream-sync label, v5 re-authoring #16583) - static-src/css/overview.css + static/index.html: v5 #16601 phone KPI fix taken, with v6's 7-tile grid (repeat(7β¦), 1799px breakpoint) retained - static-src/js/tracker-row-render.js + static/index.html: v6 finisher fields kept, v5 unauthorized_step added - src/docs/maintainer-commands.md: v6 PR-approval wording kept, v5 #16579 approved-direction text spliced into /hive decision row - src/docs/api-reference/*: ours + v5 purpose text; citations re-anchored by check-docs-citations.py --fix - src/docs/scale-envelope.md: RankActionableIssues range re-anchored manually (5314-5324) Signed-off-by: Andrew Anderson <andy@clubanderson.com>
|
@clubanderson this PR is back in your court (
Once the new head lands, CI and the review swarm re-run automatically and the PR returns to the normal path. |
|
The hive's GitHub App lacks |
β¦ v5 regions doc section unparkStageThatStays (v6 #14218) seeded the post-un-park label set with approved-direction unconditionally. v5 #16579 now derives needs-human beside the ack as held human work (stage None) rather than a 045 park, so for a kept lifecycle needs-human β where the ack is withheld β the helper misreported the hive's stage/045 as lifting and the reply said "still on the issue" instead of "stays". Only seed the ack when acknowledge is set. Fixes TestSweepLabelDrivenHiveApproveKeptNeedsHumanKeepsStage (test rest 3/4). Also re-add the "Published-checkpoint regions" prose section from v5 #16609 to api-reference/dashboard-misc.md, dropped by the row-only conflict resolution; fixes the broken #published-checkpoint-regions anchor (relative links and anchors resolve). Signed-off-by: Andrew Anderson <andy@clubanderson.com>
|
@clubanderson this PR is back in your court (
Once the new head lands, CI and the review swarm re-run automatically and the PR returns to the normal path. |
|
The hive's GitHub App lacks |
|
Thank you for your contribution! Your PR has been merged. We'd love to hear how your experience was: share feedback |
Top up v6 with the latest v5 snapshot. Replaces the failed
v6-topup.ymlrun 41762216041 (merge conflicts).Range:
origin/v6baf3e1f75βorigin/v57ac511d9c(33 commits, merge-base03b781e3b). Single plain merge commit, like #16378.Conflicts resolved (21 files)
.github/scripts/integration-governance.cjsTRUSTED_SCRIPTS_TREEre-pinned to the mergedHEAD:src/scriptstree (b3108c5a9β¦); neither parent's pin matched the merged tree..github/scripts/native-queue-policy.cjs.github/scripts/native-queue-policy.test.cjsnode --test24/24.src/cmd/hive/claimswire.goledgerMode(#16434); dropped v5's localqueue :=because v6'sgithubClaimHooksOntakes the queue as a parameter.src/crux/logscrub/basic_auth_test.go(add/add)src/crux/logscrub/handler.goScrubStringCountedapi-key counting); v5 #16470 was a backport toward v6.src/crux/logscrub/scrub_test.gosrc/docs/api-reference/health-status-events.mdcheck-docs-citations.py --fix.src/docs/api-reference/snapshots-and-style.md--fix.src/docs/operator-reference.mdsrc/docs/scale-envelope.mdRankActionableIssues(client.go:5291-5299).src/pkg/config/config.goTaskMCP+ v5ConfigGuard(#16438).src/pkg/config/testdata/agent_runtime_golden.jsonsrc/pkg/dashboard/pr_throughput_ui_test.goruns-section).src/pkg/dashboard/server.gosrc/pkg/dashboard/static-src/js/chat.jssrc/pkg/dashboard/static-src/js/refresh.jsdashboardRenderSectionwrappers; v6's earlierapplyDashboardFeatureVisibilityplacement kept (no duplicate); v6-onlyrenderRunswrapped asdashboardRenderSection('runs', ['runs-section'], β¦).src/pkg/dashboard/static/index.htmlstatic-src/permanifest.txt(byte-concat; algorithm verified to reproduce both parents' artifacts).src/pkg/github/advisory.gosrc/pkg/github/automerge/pr_approve_command_test.gosrc/pkg/github/sentinel_sweep.gowriteSentinelCommentedit-in-place helper (#16434) called with v6'smarkerBody(adopted note + digest).Judgement call to double-check
server.go: restoreds.mux.Handle("GET /", webstatic.FileServer(staticContent)). v6 carriedhttp.FileServer(http.FS(...))only through an earlier merge commit (no authored commit introduced it), i.e. a silent drop of the π fix: revalidate unversioned dashboard static files (hub /static/ and spoke stylesheets)Β #9755 revalidating file server; the v5 comment that merged in asserts the revalidating server, so I realigned the code with it.gofmt -lclean on touched Go files;check-docs-citations.pyverifies 2006 citations; no conflict markers.