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
docs(code-tidying): reword dissolve-comments' keep list #9786
Parent: #7052 (owner decision 2026-10-13: one PR per plugin).
Problem
A copy check during #7009 flagged text in this plugin that shares word runs or one-to-one structure with an upstream source. The repository's rule is our own words, with credit only on docs/upstream/ pages. Line numbers from the filing are stale; locate the passage by the finding.
Confirm each finding with fingerprint.mjs compare first. If it does not fire, record that in the PR and change nothing. If it fires, rewrite the passage in a new structure with this repository's own examples, keeping every directive and eval expectation. Make sure the upstream source has a credit row on docs/upstream/cursor-pstack.md.
Done when
gh api -H 'Accept: application/vnd.github.raw' 'repos/cursor/plugins/contents/pstack/agents/comment-sicko.md?ref=12d587dfb2' > /tmp/upstream-code-tidying-0.md then node plugins/attribution/skills/audit/scripts/fingerprint.mjs compare --local plugins/code-tidying/skills/dissolve-comments/reference/triage.md --source /tmp/upstream-code-tidying-0.md | jq .separation.fired prints false.
The PR body lists each passage with its before and after longest_span_words and containment.
Proposed work class: work-class: scoped
Basis: automated intake triage; the lanes App applied this class.
Size: Small
Complexity: Low
Uncertainty: Known
Flags: none
Understood: a copy check flagged the keep list in the code-tidying plugin's dissolve-comments triage reference as too close to an upstream source. The fix is to confirm the finding with the fingerprint tool. If it fires, reword the passage in this repository's own words and keep every directive and eval expectation. Also confirm the upstream credit row exists. Nothing appears to be missing. No likely duplicates found among the open issues.
The work-resolve-issue lane stopped on its usage limit, not on this issue. It keeps agent-ready, and work-unblock starts it again after 2026-10-25T16:00:00Z.
The work-resolve-issue lane stopped partway through this issue and needs a person.
Why, in the run's own words:
The issue's first step is to confirm the finding with fingerprint.mjs compare against cursor/plugins12d587dfb2 pstack/agents/comment-sicko.md, and to change nothing if it does not fire. This lane cannot do that: it cannot fetch the upstream file (github.com and raw.githubusercontent.com are not allowed web hosts here, and gh cannot run), and it cannot run node, so fingerprint.mjs cannot run either. Without the upstream text this run cannot tell whether the finding fires, which passage of triage.md matches, or what the before and after longest_span_words and containment values the PR body needs are. Rewriting without that check could change a passage the issue says to leave alone, so nothing was edited. Rerun this in a lane that can fetch from GitHub and run node, or run the comparison by hand and attach its output to the issue.
What the run says it did not do:
fingerprint.mjs compare of triage.md against cursor/plugins12d587dfb2 pstack/agents/comment-sicko.md (the issue's first Done-when item): did not run, because this lane cannot fetch the upstream file or run node.
The rewrite of the keep-list passage, and the PR-body table of before and after longest_span_words and containment: not done, because both depend on the comparison above.
Review, build and test steps: skipped, because there is no diff.
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 partway through this issue and needs a person.
Why, in the run's own words:
The issue makes every edit depend on node plugins/attribution/skills/audit/scripts/fingerprint.mjs compare against the upstream file fetched raw with gh api. The fix rule is 'if it does not fire, record that and change nothing; if it fires, rewrite'. This lane can't run node or gh, so neither the gating compare nor the Done-when check can run here. WebFetch returns a model summary of the upstream page, not its exact bytes, so the shingle comparison can't be reproduced by hand to tool accuracy either. Nothing was changed. A lane that can run node and gh should run the compare. The findings below suggest it will not fire on triage.md, in which case the issue closes with no change.
What the run says it did not do:
The issue's gating and Done-when check: gh api raw fetch of cursor/plugins12d587dfb2 pstack/agents/comment-sicko.md, then node plugins/attribution/skills/audit/scripts/fingerprint.mjs compare --local plugins/code-tidying/skills/dissolve-comments/reference/triage.md --source <fetched>. Not
The PR body's before and after longest_span_words and containment per passage: there are no tool figures to report because the compare could not run
Review, build/test and lint steps: skipped because nothing was changed
Closing as not planned: the copy check this issue's Fix starts with does not fire, and its rule for that case is "record that ... and change nothing".
Evidence, run 2026-10-25 against main at 5e70951: fingerprint.mjs compare --local plugins/code-tidying/skills/dissolve-comments/reference/triage.md --source <cursor/plugins@12d587dfb2:pstack/agents/comment-sicko.md, fetched raw> returned {"fired":false,"rule":"containment>=0.3||span>=15","longest_span_words":0,"containment":0,"jaccard":0}.
The upstream source already has a credit row in docs/upstream/cursor-pstack.md, so the Fix's other condition holds too.
The issue lane could not do this itself: its run stopped because it cannot run node or gh (lane outcome on #9786, run 40213739706).
Reopen it if a later check finds a structural copy the word-run compare does not measure.
Activity
github-actions commentedon Oct 24, 2026
Proposed work class: work-class: scoped
Basis: automated intake triage; the lanes App applied this class.
Size: Small
Complexity: Low
Uncertainty: Known
Flags: none
Understood: a copy check flagged the keep list in the code-tidying plugin's dissolve-comments triage reference as too close to an upstream source. The fix is to confirm the finding with the fingerprint tool. If it fires, reword the passage in this repository's own words and keep every directive and eval expectation. Also confirm the upstream credit row exists. Nothing appears to be missing. No likely duplicates found among the open issues.
Automated triage, applied by the lanes App.
melodic-automation-lanes commentedon Oct 24, 2026
The work-resolve-issue lane stopped on its usage limit, not on this issue. It keeps
agent-ready, and work-unblock starts it again after 2026-10-25T16:00:00Z.Run: https://github.com/melodic-software/claude-code-plugins/actions/runs/40140186674
melodic-automation-lanes commentedon Oct 25, 2026
The work-resolve-issue lane stopped partway through this issue and needs a person.
Why, in the run's own words:
What the run says it did not do:
Run: https://github.com/melodic-software/claude-code-plugins/actions/runs/40181681558
To retry, remove
needs-humanand addagent-readyagain.melodic-automation-lanes commentedon Oct 25, 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 25, 2026
The work-resolve-issue lane stopped partway through this issue and needs a person.
Why, in the run's own words:
What the run says it did not do:
gh apiraw fetch of cursor/plugins12d587dfb2 pstack/agents/comment-sicko.md, thennode plugins/attribution/skills/audit/scripts/fingerprint.mjs compare --local plugins/code-tidying/skills/dissolve-comments/reference/triage.md --source <fetched>. Notlongest_span_wordsandcontainmentper passage: there are no tool figures to report because the compare could not runRun: https://github.com/melodic-software/claude-code-plugins/actions/runs/40213739706
To retry, remove
needs-humanand addagent-readyagain.kyle-sexton commentedon Oct 25, 2026
Closing as not planned: the copy check this issue's Fix starts with does not fire, and its rule for that case is "record that ... and change nothing".
Evidence, run 2026-10-25 against main at 5e70951:
fingerprint.mjs compare --local plugins/code-tidying/skills/dissolve-comments/reference/triage.md --source <cursor/plugins@12d587dfb2:pstack/agents/comment-sicko.md, fetched raw>returned{"fired":false,"rule":"containment>=0.3||span>=15","longest_span_words":0,"containment":0,"jaccard":0}.The upstream source already has a credit row in
docs/upstream/cursor-pstack.md, so the Fix's other condition holds too.The issue lane could not do this itself: its run stopped because it cannot run node or gh (lane outcome on #9786, run 40213739706).
Reopen it if a later check finds a structural copy the word-run compare does not measure.