Repair director queue PR classification from PR #984 review

coilyco-ops commented

2026-07-10 09:40:51 +00:00

Member

Problem

PR ward#984 adds the director queue view for stale reservations and redispatch candidates, but review blocked it for a concrete PR classification bug:

cmd/ward/agent_director_queue.go line 231: classifyDirectorQueueOutcome checks item.Kind == backlogKindPullRequest, but classifyDirectorQueuePR stores Kind: backlogEntryKindPrefix(backlogKindPullRequest), which is the display string PR. Done PR carries therefore miss the PR branch and get close stale-open done issue.

The source issue ward#977 is consult-labeled, but this repair is now a self-contained headless correctness fix against the existing PR.

Acceptance criteria

classifyDirectorQueueOutcome correctly distinguishes PR rows from issue rows using the representation actually stored on the queue item.
Done PR carries are not classified as closeable stale-open issues.
Tests cover done PR and done issue outcomes separately.
PR #984 is updated or replaced with the repaired implementation.

## Problem PR `ward#984` adds the director queue view for stale reservations and redispatch candidates, but review blocked it for a concrete PR classification bug: ```text cmd/ward/agent_director_queue.go line 231: classifyDirectorQueueOutcome checks item.Kind == backlogKindPullRequest, but classifyDirectorQueuePR stores Kind: backlogEntryKindPrefix(backlogKindPullRequest), which is the display string PR. Done PR carries therefore miss the PR branch and get close stale-open done issue. ``` The source issue `ward#977` is consult-labeled, but this repair is now a self-contained headless correctness fix against the existing PR. ## Acceptance criteria * `classifyDirectorQueueOutcome` correctly distinguishes PR rows from issue rows using the representation actually stored on the queue item. * Done PR carries are not classified as closeable stale-open issues. * Tests cover done PR and done issue outcomes separately. * PR #984 is updated or replaced with the repaired implementation.

coilyco-ops added the

headless

P2

labels

2026-07-10 09:40:51 +00:00

coilyco-ops commented

2026-07-10 09:40:59 +00:00

Author

Member

WARD-DISPATCH: deferred ⏸

deferred details

This forwarded dispatch was deferred after the issue was already reserved.

Attempted harness: codex
Attempted run: ward agent engineer coilyco-flight-deck/ward#995 --harness codex --details Fix ward#995 by repairing the PR kind classification bug found in review of ward PR #984. Update the existing PR branch or replace it, add focused tests for PR vs issue outcomes, and close ward#995 when landed.
Container: engineer-codex-ward-995
Container created: no running engineer was observed.
Host log: /Users/kai/.ward/agent-logs/dispatch/20260710T094057Z-director-codex-zv99-coilyco-flight-deck-ward-995.log
Capacity: ward agent engineer --harness codex: global engineer limit is reached: 12 running (limit 12); wait for a run to finish or run ward agent reap for stale engineers

Retry: the issue stays queued and the director will try again when a slot opens.

— Codex, via ward agent

WARD-DISPATCH: deferred ⏸ <details><summary>deferred details</summary> This forwarded dispatch was deferred after the issue was already reserved. Attempted harness: `codex` Attempted run: `ward agent engineer coilyco-flight-deck/ward#995 --harness codex --details Fix ward#995 by repairing the PR kind classification bug found in review of ward PR #984. Update the existing PR branch or replace it, add focused tests for PR vs issue outcomes, and close ward#995 when landed.` Container: `engineer-codex-ward-995` Container created: no running engineer was observed. Host log: `/Users/kai/.ward/agent-logs/dispatch/20260710T094057Z-director-codex-zv99-coilyco-flight-deck-ward-995.log` Capacity: `ward agent engineer --harness codex: global engineer limit is reached: 12 running (limit 12); wait for a run to finish or run `ward agent reap` for stale engineers` Retry: the issue stays queued and the director will try again when a slot opens. </details>  — Codex, via `ward agent`

coilyco-ops commented

2026-07-10 17:00:27 +00:00

Author

Member

WARD-RESERVATION: held 🔒

reservation details

Holder: container engineer-claude-ward-995 on host kais-macbook-pro-2.local.

Reserved by ward agent --harness claude (reserved 2026-07-10T17:00:27Z). Concurrent ward agent runs are blocked until it finishes or the reservation goes stale (1h TTL). --force overrides.

Do not comment on or edit this issue to steer the run while it is reserved. The engineer seeded the body once at launch and never re-reads it, so a comment or edit reaches only human readers, never the running engineer. A correction goes to a new issue, dispatched fresh. That is the only channel that reaches a run in flight. Where the forge supports it, ward locks this conversation to make that a road-block rather than a convention (ward#494).

run seed context — what this run is carrying (ward#609)

Resolved: coilyco-flight-deck/ward#995 · branch issue-995 · harness claude · workflow pull-requests-and-merge
Run: engineer-claude-ward-995 · ward v0.584.0 · dispatched 2026-07-10T17:00:27Z
Comment thread: 1 included in the pre-flight read, 0 stripped (ward's own automated comments).
- included: @coilyco-ops (2026-07-10T09:40:59Z)

Static container doctrine and seed boilerplate are identical every run and omitted here (they ride ward v0.584.0).

— Claude (she/her), via ward agent

WARD-RESERVATION: held 🔒 <details><summary>reservation details</summary> Holder: container `engineer-claude-ward-995` on host `kais-macbook-pro-2.local`. Reserved by `ward agent --harness claude` (reserved 2026-07-10T17:00:27Z). Concurrent `ward agent` runs are blocked until it finishes or the reservation goes stale (1h TTL). `--force` overrides. **Do not comment on or edit this issue to steer the run while it is reserved.** The engineer seeded the body once at launch and never re-reads it, so a comment or edit reaches only human readers, never the running engineer. A correction goes to a **new issue, dispatched fresh**. That is the only channel that reaches a run in flight. Where the forge supports it, ward locks this conversation to make that a road-block rather than a convention (ward#494). <details><summary>run seed context — what this run is carrying (ward#609)</summary> - **Resolved:** `coilyco-flight-deck/ward#995` · branch `issue-995` · harness `claude` · workflow `pull-requests-and-merge` - **Run:** `engineer-claude-ward-995` · ward `v0.584.0` · dispatched `2026-07-10T17:00:27Z` - **Comment thread:** 1 included in the pre-flight read, 0 stripped (ward's own automated comments). - included: @coilyco-ops (2026-07-10T09:40:59Z) Static container doctrine and seed boilerplate are identical every run and omitted here (they ride ward v0.584.0). </details> </details>  — Claude (she/her), via `ward agent`

coilyco-ops referenced this issue from a commit

2026-07-10 17:15:00 +00:00

fix(director): classify queue PR rows by canonical backlog kind

coilyco-ops referenced this issue from a pull request that will close it,

2026-07-10 17:15:13 +00:00

Repair director queue PR classification from PR #984 review #1043

coilyco-ops referenced this issue

2026-07-10 17:15:28 +00:00

Add a director queue view for stale reservations and redispatch candidates #984

coilyco-ops commented

2026-07-10 17:17:09 +00:00

Author

Member

WARD-OUTCOME: merge-ready

details

workflow: pull-requests-and-merge; review summary: skipped - the in-container review gate was intentionally skipped by the ~/.ward/config.yaml default, so no in-container review ran; CI is the check surface.

PR: #1043 (branch issue-995), CI green (test / test success on head 15685fd5), mergeable. Supersedes and closes PR #984 (commented and closed).

What landed:

merged the issue-977 queue-view branch from PR #984 into issue-995.
repaired the review-blocking bug at the representation seam: directorQueueItem.Kind now stores the canonical backlog kind (backlogKindIssue/backlogKindPullRequest) and the PR/issue display prefix is derived only at render time, so classifyDirectorQueueOutcome's kind check compares like with like and done PR carries get wait, not close-stale-open-done-issue.
tests cover done PR and done issue outcomes separately, lock the canonical Kind on every classified item, and assert the rendered view keeps PR/issue prefixes without counting the done PR toward close-done.
also carried a hook repair: main's tip (60133a84) left a dead docs/ward-specs.md reference plus an over-cap comment block in configsource.go that failed the agentic-os pre-commit suite (source-doc-refs, code-comments) on every commit in this repo; the comment is now within cap and the monolith-constraint detail moved into docs/config-source.md.

Retrospective: the fix itself was small and felt clean - the bug was a display-string/canonical-kind mismatch, and storing the canonical kind is strictly more robust than fixing the comparison string. The surprise was main being un-committable: 60133a84 must have landed without hooks, and every commit here tripped on it, so the repair had to ride this PR. Confidence high - build/test/vet/lint all green locally (4 consecutive uncached test runs) and CI green on the PR.

Follow-ups worth an issue: main's release pipeline shows persistent publish-kdl-read / publish-kdl-write failures on recent release runs (pre-existing, unrelated to this change).

WARD-OUTCOME: merge-ready <details><summary>details</summary> workflow: pull-requests-and-merge; review summary: skipped - the in-container review gate was intentionally skipped by the ~/.ward/config.yaml default, so no in-container review ran; CI is the check surface. PR: #1043 (branch issue-995), CI green (test / test success on head 15685fd5), mergeable. Supersedes and closes PR #984 (commented and closed). What landed: * merged the issue-977 queue-view branch from PR #984 into issue-995. * repaired the review-blocking bug at the representation seam: directorQueueItem.Kind now stores the canonical backlog kind (backlogKindIssue/backlogKindPullRequest) and the PR/issue display prefix is derived only at render time, so classifyDirectorQueueOutcome's kind check compares like with like and done PR carries get wait, not close-stale-open-done-issue. * tests cover done PR and done issue outcomes separately, lock the canonical Kind on every classified item, and assert the rendered view keeps PR/issue prefixes without counting the done PR toward close-done. * also carried a hook repair: main's tip (60133a84) left a dead docs/ward-specs.md reference plus an over-cap comment block in configsource.go that failed the agentic-os pre-commit suite (source-doc-refs, code-comments) on every commit in this repo; the comment is now within cap and the monolith-constraint detail moved into docs/config-source.md. Retrospective: the fix itself was small and felt clean - the bug was a display-string/canonical-kind mismatch, and storing the canonical kind is strictly more robust than fixing the comparison string. The surprise was main being un-committable: 60133a84 must have landed without hooks, and every commit here tripped on it, so the repair had to ride this PR. Confidence high - build/test/vet/lint all green locally (4 consecutive uncached test runs) and CI green on the PR. Follow-ups worth an issue: main's release pipeline shows persistent publish-kdl-read / publish-kdl-write failures on recent release runs (pre-existing, unrelated to this change). </details>

coilyco-ops commented

2026-07-10 18:02:58 +00:00

Author

Member

WARD-RESERVATION: held 🔒

reservation details

Holder: container engineer-codex-ward-995 on host kais-macbook-pro-2.local.

Reserved by ward agent --harness codex (reserved 2026-07-10T18:02:57Z). Concurrent ward agent runs are blocked until it finishes or the reservation goes stale (1h TTL). --force overrides.

Do not comment on or edit this issue to steer the run while it is reserved. The engineer seeded the body once at launch and never re-reads it, so a comment or edit reaches only human readers, never the running engineer. A correction goes to a new issue, dispatched fresh. That is the only channel that reaches a run in flight. Where the forge supports it, ward locks this conversation to make that a road-block rather than a convention (ward#494).

run seed context — what this run is carrying (ward#609)

Resolved: coilyco-flight-deck/ward#995 · branch issue-995 · harness codex · workflow pull-requests-and-merge
Run: engineer-codex-ward-995 · ward v0.584.0 · dispatched 2026-07-10T18:02:57Z
Comment thread: 2 included in the pre-flight read, 1 stripped (ward's own automated comments).
- included: @coilyco-ops (2026-07-10T09:40:59Z), @coilyco-ops (2026-07-10T17:17:09Z)
- stripped: @coilyco-ops (2026-07-10T17:00:27Z)

Static container doctrine and seed boilerplate are identical every run and omitted here (they ride ward v0.584.0).

— Codex, via ward agent

WARD-RESERVATION: held 🔒 <details><summary>reservation details</summary> Holder: container `engineer-codex-ward-995` on host `kais-macbook-pro-2.local`. Reserved by `ward agent --harness codex` (reserved 2026-07-10T18:02:57Z). Concurrent `ward agent` runs are blocked until it finishes or the reservation goes stale (1h TTL). `--force` overrides. **Do not comment on or edit this issue to steer the run while it is reserved.** The engineer seeded the body once at launch and never re-reads it, so a comment or edit reaches only human readers, never the running engineer. A correction goes to a **new issue, dispatched fresh**. That is the only channel that reaches a run in flight. Where the forge supports it, ward locks this conversation to make that a road-block rather than a convention (ward#494). <details><summary>run seed context — what this run is carrying (ward#609)</summary> - **Resolved:** `coilyco-flight-deck/ward#995` · branch `issue-995` · harness `codex` · workflow `pull-requests-and-merge` - **Run:** `engineer-codex-ward-995` · ward `v0.584.0` · dispatched `2026-07-10T18:02:57Z` - **Comment thread:** 2 included in the pre-flight read, 1 stripped (ward's own automated comments). - included: @coilyco-ops (2026-07-10T09:40:59Z), @coilyco-ops (2026-07-10T17:17:09Z) - stripped: @coilyco-ops (2026-07-10T17:00:27Z) Static container doctrine and seed boilerplate are identical every run and omitted here (they ride ward v0.584.0). </details> </details>  — Codex, via `ward agent`

coilyco-ops commented

2026-07-10 18:09:12 +00:00

Author

Member

WARD-OUTCOME: merge-ready

details

workflow: pull-requests-and-merge; review summary: skipped - the in-container review gate was intentionally skipped by the ~/.ward/config.yaml default, so no in-container review ran; CI is green.

PR: #1043 on issue-995, mergeable, test / test (pull_request) succeeded on 8db427c1.

What landed:

merged current main into the existing PR branch.
kept the canonical-kind PR classification fix so done PR carries stay on wait.
kept the done PR and done issue tests.
resolved the configsource comment conflict from main.

Retrospective: small code fix, routine merge cleanup. Confidence high. Surprises: none.

Follow-ups: none.

WARD-OUTCOME: merge-ready <details><summary>details</summary> workflow: pull-requests-and-merge; review summary: skipped - the in-container review gate was intentionally skipped by the ~/.ward/config.yaml default, so no in-container review ran; CI is green. PR: #1043 on `issue-995`, mergeable, `test / test (pull_request)` succeeded on `8db427c1`. What landed: - merged current `main` into the existing PR branch. - kept the canonical-kind PR classification fix so done PR carries stay on `wait`. - kept the done PR and done issue tests. - resolved the configsource comment conflict from `main`. Retrospective: small code fix, routine merge cleanup. Confidence high. Surprises: none. Follow-ups: none. </details>

coilysiren closed this issue

2026-07-10 18:20:08 +00:00

coilyco-ops commented

2026-07-10 18:30:17 +00:00

Author

Member

Post-mortem — closed in the 2026-07-10 ward pileup cleanup. Root cause: a Codex director in /goal (slash-goal) mode hammered the goal criteria hard enough to spin off the rails, mass-dispatching engineers with no backpressure, producing ~30 PRs that jammed the merge queue. Prevention now owned by ward#1069 (anti-churn epic), ward#1073 (open-PR backpressure gate, default 6, broker + pre-flight), aos#450 (codex binary + director model upgrade).

PR#1043 was empty (0 commits vs main) — its director-queue-classification change had already landed. Closed as superseded/done.

**Post-mortem — closed in the 2026-07-10 ward pileup cleanup.** Root cause: a Codex director in `/goal` (slash-goal) mode hammered the goal criteria hard enough to spin off the rails, mass-dispatching engineers with no backpressure, producing ~30 PRs that jammed the merge queue. Prevention now owned by ward#1069 (anti-churn epic), ward#1073 (open-PR backpressure gate, default 6, broker + pre-flight), aos#450 (codex binary + director model upgrade). PR#1043 was empty (0 commits vs `main`) — its director-queue-classification change had already landed. Closed as superseded/done.

coilysiren referenced this issue

2026-07-14 15:27:09 +00:00

director should emit colorized, space buffered, visually distinct messages about the categories of actions its performing on startup #1283

Rows
Columns

Repair director queue PR classification from PR #984 review #995

Problem

Acceptance criteria