multi-repo-pr-check
Skill4 starsChanged 8 months agoMITView on GitHub
Triage every open pull request you own across all your repositories in one pass, reporting CI status, review state, and the next action for each. Use it at the start of a coding session, to see which PRs need attention before you open anything new; before rebasing or updating a branch, to confirm nothing upstream has changed the plan; during weekly PR triage when you maintain more than a handful of repositories; when PRs go stale because nothing surfaces them in a single place.
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 15
Run: gh search prs author:@me is:open --json repository,title,number,url
- Distinctive: line 31
Put each PR in exactly one bucket:
- Distinctive: line 43
Start with one table, sorted BLOCKED, READY, WAITING, PENDING, DRAFT:
- Very specific: line 51
If mergeable_state is dirty, say which base branch it conflicts with.
- Distinctive: line 52
Call out any PR older than 14 days explicitly, even if it is green.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
<!-- Generated from recipes/code-review/multi-repo-pr-check.md by scripts/build-skills.sh. Edit the recipe, then rerun the script. -->Multi-Repo PR Check
Triage all of my open pull requests across every repository I contribute to. This is a read-only report: do not push, merge, close, approve, or comment on anything.
1. DISCOVERY
Run: gh search prs author:@me is:open --json repository,title,number,url
2. PER-PR CHECKS
For each PR found, gather:
- Merge readiness: gh api repos/{owner}/{repo}/pulls/{number} -- read
draft,mergeable,mergeable_state,head.sha,created_at,updated_at - CI detail: gh api repos/{owner}/{repo}/commits/{head_sha}/check-runs -- name every failing or pending check, not just the rollup
- Review state: gh api repos/{owner}/{repo}/pulls/{number}/reviews -- approvals, change requests, and who asked for them
- Open feedback: gh api repos/{owner}/{repo}/pulls/{number}/comments -- treat comments newer than my last commit as unaddressed
- Age: days since the PR was opened, and days since the last push
These calls are independent and read-only, so fetch the PRs concurrently rather than one after another.
3. CLASSIFICATION
Put each PR in exactly one bucket:
- BLOCKED: failing CI, merge conflicts, or requested changes I have not answered
- READY: approved (or review not required), all checks green, mergeable
- WAITING: green and conflict-free, but review has not happened yet
- PENDING: checks still queued or running, so there is nothing to act on yet
- DRAFT: still marked as a draft
Then, on top of its bucket, flag any PR with no push and no review activity for more than 14 days as STALE.
4. OUTPUT
Start with one table, sorted BLOCKED, READY, WAITING, PENDING, DRAFT:
| Repo | # | Title | CI | Reviews | Age | Next action |
Then, in priority order, list only the PRs that need something from me. For each one, state the single concrete next step: rebase onto the base branch, fix a named failing check, reply to a specific review comment, or merge.
Rules for the report:
- Name the failing check and its failure reason. "CI red" is not an acceptable answer.
- If
mergeable_stateisdirty, say which base branch it conflicts with. - Call out any PR older than 14 days explicitly, even if it is green.
- End with a one-line count per bucket, plus how many of them are flagged STALE.
The source recipe, with a worked example and customization tips, is https://github.com/Sagargupta16/agent-recipes/blob/main/recipes/code-review/multi-repo-pr-check.md
In this repository22 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- api-docs-generatorSkill
- architecture-reviewSkill
- background-deploy-monitorSkill
- changelog-writerSkill
- ci-cd-pipelineSkill
- cjs-to-esmSkill
- class-to-hooksSkill
- coverage-gap-finderSkill
- dependency-auditSkill
- dockerfile-generatorSkill
- e2e-test-writerSkill
- generate-unit-testsSkill
- input-validationSkill
- js-to-tsSkill
- monitoring-setupSkill
- owasp-auditSkill
- performance-reviewSkill
- pr-reviewSkill
- readme-generatorSkill
- secret-scannerSkill
- upgrade-dependenciesSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

