You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
refactor(planning): one owner for the acceptance_criteria_format resolution ladder #10119
W4 of #6917. The acceptance-criteria format resolution ladder is restated in three skills, interview's SKILL.md among them. Keep one owner and point the others at it. A cut to existing skill text ships with eval evidence of no loss (AGENTS.md Code Review Rules); that spend is pre-approved within prepaid credit.
Done when
git grep -n -i 'acceptance_criteria_format' -- plugins/ shows the ladder stated in one file and pointers elsewhere.
bash plugins/planning/tests/interview-defenses.test.sh exits 0, with any pinned digest the edit moves recomputed.
The PR body carries the before and after eval results for each skill whose text was cut.
ci-status is green on the PR, and the PR body carries Refs: #6917.
Split by the owner's HITL session (2026-10-24, owner decisions C12 and Q93-Q95). Intake triage sets its fields.
Proposed work class: work-class: scoped
Basis: automated intake triage; the lanes App applied this class.
Size: Medium
Complexity: Medium
Uncertainty: Knowable
Flags: none
Blocked by: none
Understood: the acceptance_criteria_format resolution ladder is restated in three planning skills. The request is to keep it in one file and have the other two point to it. Because this cuts existing skill text, the PR needs before and after eval results. The interview-defenses test must pass, with any pinned digest recomputed. No duplicates found among the open issues.
Needs a run of claude (eval, probe or /usage), which the issue lane cannot do (pr-pipeline.yaml:289-314). The owner's HITL session runs it on the operator machine; no owner action needed.
The work-resolve-issue run that agent-ready started failed or was cancelled before it reported, or it stopped or the token broker refused it because its workflow files differ from main's. work-unblock removed and added agent-ready back to start a fresh run on main's workflow files. If this run ends the same way, it re-fires the issue again, waiting longer each time.
The work-resolve-issue run that agent-ready started failed or was cancelled before it reported, or it stopped or the token broker refused it because its workflow files differ from main's. work-unblock removed and added agent-ready back to start a fresh run on main's workflow files. If this run ends the same way, it re-fires the issue again, waiting longer each time.
The work-resolve-issue lane stopped on this issue (host-only): the run reports the work needs a host or attended run the lane cannot make.
It added host-run and removed agent-ready: a maintainer session on a host builds it, and it waits on no decision from a person.
In the run's words:
host-only: the edit is in the working tree, but two Done-when items need runs this lane cannot make. (1) A cut to existing skill text ships with before and after eval evidence for /planning:interview, /planning:prd and /planning:design (AGENTS.md Code Review Rules), and that needs a run of claude. (2) The SKILL.md Stance section is unchanged digest in plugins/planning/tests/interview-defenses.test.sh moves because the interview ladder sat inside that section. Recomputing it means running the suite or tests/repin.sh, and this lane's shell runs only its named wrappers. The justification comment is already beside the pin, so the repin needs only bash plugins/planning/tests/repin.sh 'SKILL.md Stance section is unchanged' '<why the defense still holds>' and then a passing suite run.
What the run says it did not do:
Before and after eval results for /planning:interview, /planning:prd and /planning:design, whose ladder text was cut. Collecting them needs a run of claude, which this lane cannot make. The issue comment says the owner's HITL session runs it.
bash plugins/planning/tests/interview-defenses.test.sh will fail on the 'SKILL.md Stance section is unchanged' digest, because the interview ladder sat inside that section. The digest was not recomputed: this lane cannot run the suite, repin.sh or a hash tool. The justification comment is in place.
Tidy skipped: the change spans two areas (plugins/planning and docs/conventions) and touches the shared path docs/conventions/, both outside the tidy cutoffs.
check-affected skipped changed-skills (skill-quality over the three edited SKILL.md files) and validate-plugins sub-checks because they run the checked tree's code. Left to the PR's CI.
ci-status green on the PR, and Refs: #6917 in the PR body: left to the PR's CI and the lane that opens the PR.
Where the owner lives is this run's choice: plugins/planning/context/, beside self-answer.md and problem-framing.md, which several planning skills already share. The issues do not name a location. design was included because its copy was the same ladder for the diagram_dialect keys, which the issue'
Untracked files .bash_profile, .bashrc, .gitconfig, .idea, .mcp.json, .profile, .ripgreprc, .vscode, .zprofile and .zshrc appear in git status. This run did not create them and left them alone; they are not part of the change.
Activity
github-actions commentedon Oct 25, 2026
Proposed work class: work-class: scoped
Basis: automated intake triage; the lanes App applied this class.
Size: Medium
Complexity: Medium
Uncertainty: Knowable
Flags: none
Blocked by: none
Understood: the acceptance_criteria_format resolution ladder is restated in three planning skills. The request is to keep it in one file and have the other two point to it. Because this cuts existing skill text, the PR needs before and after eval results. The interview-defenses test must pass, with any pinned digest recomputed. No duplicates found among the open issues.
Automated triage, applied by the lanes App.
kyle-sexton commentedon Oct 25, 2026
Needs a run of
claude(eval, probe or /usage), which the issue lane cannot do (pr-pipeline.yaml:289-314). The owner's HITL session runs it on the operator machine; no owner action needed.melodic-automation-lanes commentedon Oct 27, 2026
The work-resolve-issue run that
agent-readystarted failed or was cancelled before it reported, or it stopped or the token broker refused it because its workflow files differ from main's. work-unblock removed and addedagent-readyback to start a fresh run on main's workflow files. If this run ends the same way, it re-fires the issue again, waiting longer each time.melodic-automation-lanes commentedon Oct 28, 2026
The work-resolve-issue run that
agent-readystarted failed or was cancelled before it reported, or it stopped or the token broker refused it because its workflow files differ from main's. work-unblock removed and addedagent-readyback to start a fresh run on main's workflow files. If this run ends the same way, it re-fires the issue again, waiting longer each time.melodic-automation-lanes commentedon Oct 28, 2026
The work-resolve-issue lane stopped on this issue (
host-only): the run reports the work needs a host or attended run the lane cannot make.It added
host-runand removedagent-ready: a maintainer session on a host builds it, and it waits on no decision from a person.In the run's words:
What the run says it did not do:
claude, which this lane cannot make. The issue comment says the owner's HITL session runs it.Refs: #6917in the PR body: left to the PR's CI and the lane that opens the PR.Run: https://github.com/melodic-software/claude-code-plugins/actions/runs/40623280765
To hand it back to the lane, remove
host-runand addagent-readyagain.