Conversation
Codifies the pass a maintainer makes repeatedly: gather state for every open PR, bucket each one into merge-now / approve+merge / needs-triage / blocked, and act only on what is confirmed. Two `gh pr list` calls cover the queue -- one for state, one for a check rollup collapsed to only non-green results -- which avoids the per-PR loop that trips gh's spurious 401. Records the judgement rules that decide most of the queue: BLOCKED + REVIEW_REQUIRED means "needs one approval" rather than "broken", self-authored PRs can never be self-approved, carried-over warnings are not regressions, and a held bump is often superseded rather than fixed. Routes anything red, or any bump being approved on substance, to analyze-action-pr rather than duplicating its failure taxonomy. Generated-by: Claude Opus 5
Adds an Agent Skills section covering analyze-action-pr and sweep-open-prs, with a TOC entry. Points readers at Batch-Reviewing Dependabot PRs as the non-agent equivalent of the queue sweep. analyze-action-pr had never been documented; it is included here rather than left invisible now that the section exists. Generated-by: Claude Opus 5
|
This one adds a SKILL - reflecting what I have been doing recently - swiping through all PRs with my agent and acting appropriately. With this skill ANYONYE can do what I did regularly by just asking the agent " "sweep throuh open PRs". That's it - everything will happen automatically then, you will be asked to confirm the actions when they are safe, or asked to additionally review the PRs when there are doubts. This should allow anyone to spend few minutes (!) a day - to keep the repo in a good shape. @dfoulks1 @dave2wave @ppkarwasz -> I would love if you would like to try it and see how much time it saves. |
|
Also @raboof -> With that SKIL I was able to sweep through 14 open PRs in lest than 10 minutes of elapsed time, with analysis of what they do, checking if they are sound, if they have no suspicious changes (this all on top of the deterministic analysis that we run in the PRs). I highly recommend merging that one and starting using it by everyone from time to time. That will help us to avoid SPOFs and delay. I even have an idea that when we have access to This would be fantastic workflow - we would just get a single place where regularly. we will see:
And any of us would be able to "Approve" all of them. |
A CheckRun carries conclusion: null until status reaches COMPLETED, and its state is always null -- state belongs to the StatusContext shape. Reaching straight for (.conclusion // .state) printed a bare "name=" for every check still running, which reads as a finding rather than "ask again in a minute". It misreported a PR whose ten checks were simply mid-flight. Read status first and report QUEUED / IN_PROGRESS, and record alongside the UNKNOWN mergeability case that neither is a verdict. Generated-by: Claude Opus 5
Adds a
sweep-open-prsskill for the pass over the whole open PR queue: gather state for every open PR, bucket each one, act only on what is confirmed.gh pr listcalls cover the queue - one for state, one for a check rollup collapsed to only non-green results. Avoids the per-PR loop that tripsgh's spurious 401.BLOCKED+REVIEW_REQUIREDmeans "needs one approval" rather than "broken"; a self-authored PR can never be self-approved; carried-over warnings are not regressions; a held bump is often superseded rather than fixed.analyze-action-prinstead of duplicating its failure taxonomy.README.mdunder a new Agent Skills section, with a TOC entry.analyze-action-prhad never been documented; it is included rather than left invisible now that the section exists.Test plan:
prek run --files README.md .claude/skills/sweep-open-prs/SKILL.mdpasses. Frontmatter is 363 chars against the 1024 limit, with the two required fields. Both README links resolve to files in the tree and the#batch-reviewing-dependabot-prsanchor matches a real heading. Everyghinvocation in the skill was run against this repo's live queue while drafting it.🤖 Generated with Claude Code