[clubanderson]

Fixes #14572

What

Draft PRs on the dashboard Hive activity monitor (e.g. #14427 / #14428 while isDraft=true + prow do-not-merge/work-in-progress) rendered as a bare "Open · since 18h" row with no status pill, which read as a stuck item. The hive's review and merge lanes skip drafts on purpose, so the wait is the author's — the row should say so.

Changes

• Payload (stage_tracker.go): the draft modifier now fires when the PR snapshot's draft bit is set or the PR carries prow's do-not-merge/work-in-progress label. That WIP label is read as a draft on the tracker rather than a hold (it used to match holdLabels' do-not-merge/* prefix), so a WIP-only PR reads draft; a real hold/do-not-merge/hold still leads (hold +1).
• Frontend (static/index.html): muted draft pill (same .ham-badge treatment as hold, not red) with tooltip "Draft PR — hive review and merge lanes skip drafts until the author marks it ready for review."; Open cell reads "Draft · since 18h" with a ✎ glyph in the draft tint and never takes the stale red tint; headline/tooltip say "Open · draft". Hold and blocked still win the cell.
• Tests: new table-driven TestStageTrackerModifiers_Draft (draft bit; WIP label only; neither; draft+hold → ["hold","draft"]; WIP+hold; hold only; draft+blocked). Node fixture assertions in TestStageTrackerRows for the Open cell, segment class, pill HTML/tooltip, hold +1 fold, stage-50 draft, and current-stage HTML. No time.Sleep.
• Docs: src/docs/dashboard-sections.md activity-monitor paragraph. Changelog changelog.d/added-tracker-draft-pill.md.

Webhook plumbing (ready_for_review / converted_to_draft) already refreshes Draft via patchPullRequest / patchWorkPRAny; no change needed there.

[kubestellar-prow]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please assign hanthor for approval. For more information see the Code Review Process.

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

Details

Needs approval from an approver in each of these files:

• OWNERS

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

[hivecommons-hive]

@clubanderson is auto-promoted for contributor-lane merging on this PR: the last 5 merged PRs by this author in this repository have no revert or sentinel-alert evidence. CI, reviewer verdict, diff fences, soak, holds and branch protection still apply.

[github-actions]

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
#14821	🐛 fix: preserve supersession decisions across stage reconciliation	same files	src/pkg/dashboard/stage_tracker.go src/pkg/dashboard/static/index.html
#14807	✨ dashboard: stream Job pod logs in the terminal pane for sandbox.runtime: job agents	same files	src/pkg/dashboard/static/index.html

[hivecommons-hive]

@clubanderson this PR is held at stage/071-needs-author and the hive will not merge it yet.

Failing checks:

• test (dashboard 1/4)
• test (dashboard shuffle 2/3)

Next: the hive does not push to your branch. Push a new head (or reply here); CI and the review swarm re-run automatically and the PR returns to the normal path.

[hivecommons-hive]

@clubanderson this PR is held at stage/071-needs-author and the hive will not merge it yet.

Why:

• the branch conflicts with the base and needs a rebase

Next: the hive does not push to your branch. Push a new head (or reply here); CI and the review swarm re-run automatically and the PR returns to the normal path.

[hivecommons-hive]

Read the Go change against the tree at 4e6ec94 plus the JS it feeds; the logic looks correct to me. One blocker that isn't the code: GitHub reports this CONFLICTING against v5 (conflicts-with-open-pr, DIRTY) — #14571 landed on v5 after this branched and touches stage_tracker.go and stage_tracker_ui_test.go, so it needs a rebase before the merge lane can take it. CI on the current head is otherwise green.

What I verified:

• src/pkg/dashboard/stage_tracker.go:642-651 — the label loop lower-cases before the switch, so the prowWIPLabel case and the EqualFold in the DeleteFunc filter agree on case/whitespace; holdLabels (bands.go:571) matches do-not-merge/* by prefix, and the delete removes only the exact WIP label, so do-not-merge/hold still sets modHold. The output order at stage_tracker.go:665 keeps hold ahead of draft, which is what the hold +1 fold in the UI test relies on.
• static/index.html:13906-13920 — openDraft is gated on !openHold && !openBlocked, the stale tint is suppressed only for the draft word, and openSince still computes for drafts so the age shows. --band-draft exists in both themes (index.html:87,190), so .ham-open-draft resolves.
• PR body matches the diff (payload, frontend, tests, docs, changelog).

One cosmetic note, not a defect: stageTrackerHoldLabels (index.html:13829-13836) still matches do-not-merge/*, so for a PR carrying both do-not-merge/hold and do-not-merge/work-in-progress, the headline at index.html:13788 will read Open · on hold (do-not-merge/hold, do-not-merge/work-in-progress). WIP-only rows are unaffected since mods no longer contains hold. Fine to leave or filter the WIP label there too for consistency with the Go side.

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

[hivecommons-hive]

Read the modifier path and its caller, not just the diff. This looks correct to me.

• src/pkg/dashboard/stage_tracker.go:642-650 — do-not-merge/work-in-progress now sets draft and is filtered out of holdLabels(...) before the hold modifier fires. Verified the body's claim that it used to read as a hold: src/pkg/dashboard/bands.go:571 matches any do-not-merge/ prefix, so a WIP-only PR did get hold before this change. A real hold / do-not-merge/hold still survives the DeleteFunc and leads the pill (hold +1), as the body says.
• static/index.html: openDraft is only true when n === 0 && !openHold && !openBlocked, and stale is masked by !openDraft — matches the stated "hold and blocked still win the cell / never tints stale".
• Docs and changelog fragment (root changelog.d/, correct location) match the behaviour.

One thing blocking it: the branch no longer merges into v5 — src/pkg/dashboard/static/index.html conflicts with the head that landed via #14571 (I reproduced it with a local merge). Needs a rebase; nothing else from me.

Open question (not a defect, out of this PR's scope): bands.go:571 still classifies a WIP-only draft as held for the Overview band / heldReason, so the band and the tracker now disagree on what WIP means. Worth a follow-up if the band should match.

— hive: agent=reviewer-2 backend=copilot model=claude-fable-5.1 copilot=1.0.88

[clubanderson]

Rebased onto v5 after #14571 — stageTrackerChipsHtml now keeps both the ham-mod-handled class for the muted "stage ↓" regressed pill and the draft tooltip.

On bands.go holdLabels treating a WIP-only PR as held: I looked at it and left it alone deliberately. holdLabels feeds prBand (the waiting band), the ⏸ held signal, heldReason's "On hold — label …" text, api_hive_advice's Held, and the issue hold check, so dropping do-not-merge/work-in-progress there would change band/signal output for every WIP PR (a WIP-labelled non-draft PR would fall through to open). That's a wider behaviour change than this PR should carry; this change scopes the "WIP reads as draft, not hold" rule to the tracker's modifier derivation. Happy to follow up separately if we want the bands to agree.

[hivecommons-hive]

@clubanderson this PR is held at stage/071-needs-author and the hive will not merge it yet.

Why:

• the branch conflicts with the base and needs a rebase

Next: the hive does not push to your branch. Push a new head (or reply here); CI and the review swarm re-run automatically and the PR returns to the normal path.

[hivecommons-hive]

Read stage_tracker.go, bands.go and the index.html segment/pill code at this head. The tracker change looks correct to me: stageTrackerModifiers (stage_tracker.go:709-718) now sets draft on prow's do-not-merge/work-in-progress and strips only that label from the holdLabels count, so a draft PR wearing nothing else reads draft instead of hold; the JS orders openDraft after hold and blocked and excludes drafts from the stale tint, matching the Go priority list.

One low-severity note, not a blocker and arguably out of #14572's scope: the same WIP label still counts as a hold everywhere else that calls holdLabels — bands.go:209 puts the PR in the waiting band (so the draft band at bands.go:224 stays unreachable for any prow-labelled draft), bands.go:242 adds the ⏸ held signal, and api_hive_advice.go:184 reports Held: true. After this lands a maintainer sees "Draft" in the tracker and "⏸ held / waiting" in the band for the same PR. If you want the two views to agree, the WIP exclusion probably belongs inside holdLabels itself (or a shared helper) rather than at the one tracker call site; happy to leave that to a follow-up.

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

[hivecommons-hive]

Evidence bundle `#14573e54c67f` sha256:08be69feedcc2a5fd12e3636c347ea86c99c719122a98f7514f12469ad67719c — download from the dashboard or `hivectl review evidence #14573`.

[hivecommons-hive]

merged by Hive trusted-author auto-merge: author @clubanderson holds role owner and GitHub permission admin on repo; CI green at e54c67f.

[github-actions]

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

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

[hivecommons-hive]

Read the Go and the tracker JS against v5; looks correct to me.

• src/pkg/dashboard/stage_tracker.go:777: do-not-merge/work-in-progress now sets draft and is filtered out of holdLabels() before the modHold check, so a PR wearing only prow's WIP label reads draft rather than hold. Any other do-not-merge/* or hold label still sets modHold (bands.go:571), so a real maintainer hold on a draft PR is not masked. modDraft already existed in stageTrackerModifierOrder and stageTrackerSunkMods, so ordering and sink behaviour are unchanged.
• static/index.html: the Draft word / ✎ chevron apply only when n === 0 && state === 'active' && !openHold && !openBlocked, and the stale tint is suppressed only in that same case, so a blocked or held Open cell keeps its existing stale signal.

Matches the scope of #14572 as the body states.

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