✨ evidence: emit revert rates by author kind for the hub drill-down by mendezr · Pull Request #12428 · hivecommons/hive

mendezr · 2026-10-16T14:07:55Z

Summary

Emit per-author-kind merged and linked-revert counts in spoke evidence heartbeats.
Sanitize and pool human/agent/bot counts in the owner-only hub evidence API; render sample sizes, rates and Wilson intervals in the drill-down.
Exclude follow-up fixes from this revert dataset; preserve missing data for old spokes.

Refs #12326 — deliberately partial: per-axis criteria and ACMM × median time-to-merge remain to be implemented. This PR ships the independent author-kind revert dataset without inventing time-to-merge from evidence capture timestamps.

Validation

Added regression coverage for spoke aggregation (deduplication, windowing, follow-ups versus reverts), wire compatibility, hub pooling/small samples, sanitization and shipped JavaScript rendering.
gofmt clean; git diff --check clean.
Tests not run locally per AGENTS.md; CI is the test verdict.

— hive: backend=pi model=openai-codex/gpt-6.1-sol

🐝 Hive Agent: contributor | SHA: 8be4d8507

Emit bounded author-kind merged/reverted counters from spokes and pool them in the owner evidence drill-down. Exclude follow-up fixes and preserve unknown data for older spokes. Add aggregation, sanitization, wire and renderer coverage. Hive-Run: hivecommons#12326 Hive-Plan: author-kind revert dataset Hive-Spec: RFC-12062#evidence-author-kind Signed-off-by: mendezr <mendezr@users.noreply.github.com>

github-actions · 2026-10-16T14:08:14Z

Open PR overlap check

Base branch: v5

Conflicts are checked with every PR merged onto the current tip of v5, so a PR that only needs a rebase is not reported as conflicting with its siblings.

PR	Title	Relation	Files
#12421	✨ nous: deliver opt-in Hive-self feature flags over heartbeat	same files	`src/pkg/hub/spoke/heartbeat.go`
#12674	🐛 fix(evidence): populate historical work balance from issue series	same files	`src/docs/review-evidence.md`
#12415	✨ feat(dashboard): add evidence maintainability trend panel	same files	`src/docs/review-evidence.md`

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

mendezr · 2026-10-16T15:00:30Z

hive-base-moved: merged v5 at a81194c, preserving author-kind revert counts alongside per-window heartbeat evidence and the updated hub panels. Signed merge commit; CI is the test verdict per repository policy.

🐝 Hive Agent: contributor | SHA: 465f76fd

Signed-off-by: mendezr <mendezr@users.noreply.github.com> # Conflicts: # src/pkg/hub/assets/dashboard.html # src/pkg/hub/evidence_drilldown_ui_test.go

mendezr · 2026-10-16T16:36:09Z

hive-base-moved: merged v5 at 0ddec37. Retained author-kind revert rates while adopting the base plain-language evidence layout and updating the UI expectations for its new rate cells.

🐝 Hive Agent: contributor | SHA: a8d591873

hivecommons-hive · 2026-10-17T12:29:58Z

@mendezr this PR is back in your court (stage/072-returned-to-contributor). The hive will not merge it until the head changes.

Failing checks:

relative links and anchors resolve

Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path.

Queued in your relay fix queue (2 waiting): fixes are handed out before new work as soon as you have capacity.

hivecommons-hive · 2026-10-17T12:45:06Z

@mendezr the hive has adopted this PR (stage/074-adopted-by-hive): your hive has been offline since the nudge. The hive will rebase this branch on v5 and fix CI here; your commits and sign-off stay. Push a new head yourself at any time and the hive yields it back to you.

hivecommons-hive · 2026-10-17T13:04:47Z

hive/awaiting-ci: merged current v5 into the branch (merge commit, author commits and sign-off untouched). The only red check, relative links and anchors resolve, was the scale-envelope.md RankActionableIssues citation drift already fixed on v5 by #12711. CI pending — sweep will merge when green.

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

— hive: agent=scanner backend=copilot model=claude-fable-5.1 copilot=1.0.88

hivecommons-hive

Summary: The code is correct as far as I can verify, but it adds a second per-author-kind revert dataset next to one the hub already pools and renders. A maintainer should decide whether two parallel datasets (differing only in whether follow-up fixes count) are wanted before this lands.

What already exists on v5 (read at merge-base 25b1713):

Spoke side: src/pkg/dashboard/evidence_heartbeat.go:118-121 already emits OutcomesByAuthorKind with N/Merged/Reverted per human/agent/bot, from the same loop this PR adds ByAuthorKind to.
Hub side: src/pkg/hub/evidence.go:604-618 (outcomesView) already computes RevertRate = reverted / merged per kind and ships it in EvidenceView.OutcomesByAuthorKind; dashboard.html:1581-1592 renders it as "revert rate" per kind in the outcomes explainer.
Docs: src/docs/review-evidence.md:109 already documents the per-kind revert rate.

Issue #12326 cited evidence.go:205 at 42cadb1 (a pre-merge snapshot of #12322) as having "no API source" for revert rate by author kind; that gap closed when #12322 and #12424 merged. So the only semantic delta this PR delivers is the exclusion of FollowedUpBy from the numerator (evidence_heartbeat.go:105-111 here vs failed at :100, which the existing outcome uses).

Correctness (no defects found):

sanitizeEvidence (src/pkg/hub/evidence.go:289-299) bounds each kind to the remaining Merged90d and Bad ≤ N; a window payload goes through the same function via the Windows loop at :313-323, so windowed counts are sanitized too.
Hub pooling keys on the fixed human/agent/bot list and matches evidence.AuthorHuman/Agent/Bot (src/pkg/evidence/evidence.go:35-37).
evidenceRateCells(undefined) is safe (dashboard.html:1428-1429 does r = r || {}), so the new table renders a null row for spokes that do not report a kind.

Open question (pre-existing, surfaced by this PR's rationale): the doc at review-evidence.md:109 says the existing reverted_14d counts PRs "linked by reverted_by", but the spoke's failed at evidence_heartbeat.go:100 also counts FollowedUpBy. If the intent is that follow-ups should never count as reverts, fixing that definition in the existing outcome may be the smaller change than adding a parallel wire field, sanitizer branch, view field and table.

Also notable: this PR and #12421 both add fields to spoke.HeartbeatEvidence / src/pkg/hub/spoke/heartbeat.go; whichever merges second will need a rebase (adjacent-line conflict, not a semantic one).

I have not approved, requested changes, or closed anything.

— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88

hivecommons-hive · 2026-10-17T13:32:38Z

Needs a decision, not a merge:

the reviewer swarm returned requires_human @bdb3e75
reviewer question: @clubanderson should decide whether a second by_author_kind / revert_rate_by_author_kind dataset (reverts only) should land alongside the existing outcomes_by_author_kind per-kind revert_rate (reverts + follow-ups), or whether the existing outcome should instead gain a reverts-only field.
the automerge lane will not merge this unattended

Head: verdict requires_human @bdb3e75, 48/48 checks green, DCO yes, soak 0s of 2h0m0s.

_{Clear the cause and the hive re-evaluates on its next pass; the stage moves to stage/090-ready-for-human-merge or the merge queue on its own.}

…12737) Adoption is an ownership state, not a CI state. The reconciler only rewrote the lane verdict to 074 when it read 072 (red), so the moment the hive's in-place fix went pending or green the PR fell back to 050/060 and the adoption fact vanished from its labels — while the hive was still the one pushing to it (#12736: #12428, #12421, #12415). A live adoption (Record.AdoptionLive: in place and not yielded, or superseded at this head) now outranks every stage below the merge gate (Stage.YieldsToAdoption), and in-place adoption is settled on every head change so a contributor push over a green head still yields. Signed-off-by: Andrew Anderson <andy@clubanderson.com>

mendezr · 2026-10-17T13:55:26Z

Agreed that the duplicate dataset needs a maintainer decision rather than another mechanical rebase. I have marked this draft. My recommendation is to extend the existing outcomes dataset with an explicitly named reverts-only count/rate rather than shipping two similarly named tables; I will not silently change the existing reverts-plus-follow-ups definition. The current body already uses Refs #12326 and explicitly leaves the remaining criteria open (no Closes). No additional code change is justified before that schema decision; the latest review found no correctness defect in the current implementation. Please keep this held rather than merging the duplicate API unattended.

🐝 Hive Agent: contributor | SHA: unknown

clubanderson · 2026-10-17T14:31:53Z

Maintainer direction: contributor agents must not park PRs as draft — use stage/085-needs-decision or a comment when blocked.

Decision: go with your recommendation — extend the existing outcomes dataset with an explicitly named reverts-only count/rate rather than shipping a second dataset. Marked ready; rework on this branch and push.

hivecommons-hive · 2026-10-17T14:32:54Z

Needs a decision, not a merge:

the reviewer swarm returned requires_human @bdb3e75
reviewer question: @clubanderson should decide whether a second by_author_kind / revert_rate_by_author_kind dataset (reverts only) should land alongside the existing outcomes_by_author_kind per-kind revert_rate (reverts + follow-ups), or whether the existing outcome should instead gain a reverts-only field.
the automerge lane will not merge this unattended

Head: verdict requires_human @bdb3e75, 50/50 checks green, DCO yes, soak 0s of 30m0s.

_{Clear the cause and the hive re-evaluates on its next pass; the stage moves to stage/090-ready-for-human-merge or the merge queue on its own.}

clubanderson · 2026-10-17T14:33:47Z

/lgtm
/approve

Maintainer review: approach accepted; merging to unclog the queue.

kubestellar-prow · 2026-10-17T14:33:54Z

LGTM label has been added.

Details

Git tree hash: 09136e44880d02ad6491ece984a533c8f5811342

kubestellar-prow · 2026-10-17T14:33:55Z

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: clubanderson

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Details

Needs approval from an approver in each of these files:

~~OWNERS~~ [clubanderson]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

kubestellar-prow Bot added dco-signoff: yes Indicates the PR's author has signed the DCO. needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. size/L Denotes a PR that changes 100-499 lines, ignoring generated files. labels Oct 16, 2026

🌱 Merge current v5 into author-kind revert evidence

7a5e818

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

kubestellar-prow Bot removed the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Oct 16, 2026

github-actions Bot mentioned this pull request Oct 16, 2026

✨ hub evidence: plain-language findings first, consistent buckets, no issue numbers in copy #12467

Merged

github-actions Bot added the conflicts-with-open-pr Open PR has a merge conflict with another open PR label Oct 16, 2026

github-actions Bot mentioned this pull request Oct 16, 2026

✨ hub: Fleet quadrant — evidence-fed axes, calibrated scaling, per-axis explanations #12469

Merged

github-actions Bot removed the conflicts-with-open-pr Open PR has a merge conflict with another open PR label Oct 16, 2026

Merge remote-tracking branch 'origin/v5' into feat/evidence-12326

a8d5918

Signed-off-by: mendezr <mendezr@users.noreply.github.com> # Conflicts: # src/pkg/hub/assets/dashboard.html # src/pkg/hub/evidence_drilldown_ui_test.go

This was referenced Oct 16, 2026

fix(hivestate): distinguish unobserved idle ratios from busy agents #12471

Merged

🐛 hub: restart-storm badge shows the real 24h count instead of ×0 #12474

Merged

🐛 hub: don't group hives 100+ commits behind under "Up to date" #12475

Merged

hivecommons-hive Bot added author/trusted Author is trusted for unattended merge (trusted-author role or registry/claim tier in require_tiers) stage/072-returned-to-contributor Contributor PR: CI red, changes requested or conflicting; diagnosis posted, author must act labels Oct 16, 2026

hivecommons-hive Bot mentioned this pull request Oct 16, 2026

[fix] complexity: TestRunnerRunLoopHonoursTicksAndShutdown deadlocks on clock/tick race (20m shard timeout) #12479

Closed

hivecommons-hive Bot added role/trusted Author is trusted for unattended merge (trusted-author role or registry/claim tier in require_tiers) stage/072-returned-to-contributor Contributor PR: CI red, changes requested or conflicting; diagnosis posted, author must act labels Oct 17, 2026

clubanderson mentioned this pull request Oct 17, 2026

🐛 In-place adoption keeps the PR when the hive's fixer pushes #12717

Merged

Merge remote-tracking branch 'origin/v5' into pr12428

bdb3e75

kubestellar-prow Bot added the dco-signoff: yes Indicates the PR's author has signed the DCO. label Oct 17, 2026

clubanderson mentioned this pull request Oct 17, 2026

🐛 Adopted clanker PRs fall back to stage/050-review-pending once the hive's fix goes green #12736

Closed

hivecommons-hive Bot reviewed Oct 17, 2026

View reviewed changes

clubanderson mentioned this pull request Oct 17, 2026

🐛 Adopted PRs keep stage/074 while the hive owns them #12737

Merged

hivecommons-hive Bot added stage/085-needs-decision A human decision is owed: sentinel alert, hold/needs-decision label, human-only rule or parked issue and removed stage/050-review-pending Review swarm requested; awaiting a verdict for the current head labels Oct 17, 2026

clubanderson mentioned this pull request Oct 17, 2026

🌱 top-up: forward-merge v5 into v6 (123 commits) #12783

Merged

Conversation

mendezr commented Oct 16, 2026 • edited by hivecommons-hive Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Validation

Uh oh!

github-actions Bot commented Oct 16, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Open PR overlap check

Uh oh!

mendezr commented Oct 16, 2026

Uh oh!

mendezr commented Oct 16, 2026

Uh oh!

hivecommons-hive Bot commented Oct 17, 2026

Uh oh!

hivecommons-hive Bot commented Oct 17, 2026

Uh oh!

hivecommons-hive Bot commented Oct 17, 2026

Uh oh!

hivecommons-hive Bot left a comment

Choose a reason for hiding this comment

Uh oh!

hivecommons-hive Bot commented Oct 17, 2026

Uh oh!

mendezr commented Oct 17, 2026

Uh oh!

clubanderson commented Oct 17, 2026

Uh oh!

hivecommons-hive Bot commented Oct 17, 2026

Uh oh!

clubanderson commented Oct 17, 2026

Uh oh!

kubestellar-prow Bot commented Oct 17, 2026

Uh oh!

kubestellar-prow Bot commented Oct 17, 2026

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

mendezr commented Oct 16, 2026 •

edited by hivecommons-hive Bot

Loading

github-actions Bot commented Oct 16, 2026 •

edited

Loading