pr-write
Skill16 starsChanged 5 months agoMITView on GitHub
Create or update a concise GitHub pull request for an already verified code change. Use when the user asks to open or file a PR, or invokes pr-write. Do not publish a PR when the user only asks for draft wording.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- Eatgrapes/WinIslandtheir own version, first seen Nov 9, 2026
- GA14-hub/claude-code-kit-jatheir own version, first seen Feb 13, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- Eatgrapes/WinIsland★ 405same name
- GA14-hub/claude-code-kit-ja★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository94 other files it gives its agents.
- agent-friction-ledgerSkill
- agent-learning-layer-triageSkill
- agent-operations-contractSkill
- agents-best-practicesSkill
- agent-tool-contract-reviewSkill
- anti-slop-reviewSkill
- architecture-deepening-reviewSkill
- assumption-excavationSkill
- authorized-security-routerSkill
- book-to-skillSkill
- bounded-evaluationSkill
- bounded-growth-loopSkill
- bounded-orchestration-contractSkill
- brief-first-executionSkill
- browser-verificationSkill
- bug-evidence-protocolSkill
- codex-review-loopSkill
- compaction-survivalSkill
- complexity-optimizerSkill
- confidence-fragility-reviewSkill
- context-degradation-reviewSkill
- corporate-surfaceSkill
- delivery-baseline-auditSkill
- delivery-preflight-gateSkill
- design-critique-polishSkill
- designmd-brand-kitSkill
- design-register-bootstrapSkill
- diagnoseSkill
- doc-grounded-grillingSkill
- doc-to-lora-evaluatorSkill
- durable-context-maintenanceSkill
- dynamic-workflow-packetsSkill
- editorial-surfaceSkill
- evidence-ledger-researchSkill
- filesystem-context-disciplineSkill
- fluid-interaction-reviewSkill
- frontend-product-builderSkill
- frontend-taste-layerSkill
- git-native-context-contractSkill
- goal-loop-designerSkill
- graph-guided-code-readingSkill
- handoffSkill
- happy-handoffSkill
- html-brief-artifactSkill
- html-diagram-artifactSkill
- hypothesis-diversificationSkill
- idea-to-ship-gatesSkill
- knowledge-base-enrichmentSkill
- lean-context-layoutSkill
- local-inference-tuningSkill
- loop-hardening-contractSkill
- loopops-protocolSkill
- loop-readiness-reviewSkill
- lottie-motion-builderSkill
- minimal-diff-builderSkill
- motion-review-gateSkill
- overengineering-reviewSkill
- personal-workspace-routerSkill
- phase-spec-executionSkill
- pipeline-readiness-gateSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 6 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
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
- Distinctive: line 17
Before updating an existing PR branch, inspect auto-merge and merge-queue state.
- Distinctive: line 21
Read the repository template and recent titles for conventions.
- Distinctive: line 23
If the authorized creation requires committing and pushing this task's change, do so on a task-owned non-default branch after verification.
- Distinctive: line 35
Use accurate model/harness attribution only when repository policy requires it.
- Distinctive: line 37
Check the whole draft and attachments for secrets, private local paths, customer data and private transcript excerpts.
- Very specific: line 41
Use the connected GitHub tools when sufficient, otherwise authenticated gh.
- Distinctive: line 45
Verify the returned PR URL and head/base.
- Very specific: line 47
Do not post unsolicited comments, resolve reviews, enable auto-merge, merge, create a release, or start indefinite monitoring.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
PR Write
Explain the concrete problem and resulting behavior. Match the repository's PR template and conventions. A request for text alone authorizes drafting only.
Verify the final scope
Before updating an existing PR branch, inspect auto-merge and merge-queue state. If either can merge it automatically, stop updates unless changing that state is authorized. Never disable it silently or promise no merge while it remains enabled.
Resolve the repository, current branch, actual target base and complete diff, including relevant untracked files. Check for an existing PR before creating one. Never stage, commit or include unrelated user work.
Read the repository template and recent titles for conventions. Confirm intended behavior with appropriate evidence, and distinguish pre-existing failures. Do not claim tests ran when they did not.
If the authorized creation requires committing and pushing this task's change, do so on a task-owned non-default branch after verification. Inspect exactly what is staged first. Do not rewrite unrelated commits or push the default branch.
Write
Lead the title and description with the problem and resulting behavior, not an inventory of files. Use a concrete before/after example where it helps. For small changes, one or two sentences and a validation line are sufficient.
Include:
- why the change matters and what now happens;
- verification actually observed, with material gaps;
- source PR/contributor credit when continuing another person's work.
Use accurate model/harness attribution only when repository policy requires it. Follow the user's language and formatting preferences; do not force an emoji signature. Never invent a model name.
Check the whole draft and attachments for secrets, private local paths, customer data and private transcript excerpts. Do not upload private logs to prove a public claim.
Publish only the requested artifact
Use the connected GitHub tools when sufficient, otherwise authenticated gh. Pass multiline bodies as structured text, or save a UTF-8 body file and use --body-file; never interpolate body text into shell code.
Honor an explicit draft request. Otherwise create a normal PR when verification is sufficient and the repository expects it; use a draft with an honest blocker if the task calls for preserving incomplete work. For a wording-only request, return the text without creating or changing anything.
Verify the returned PR URL and head/base. Update the existing PR rather than creating a duplicate. Attach the PR to the current Codex task if that host exposes an attachment tool.
Do not post unsolicited comments, resolve reviews, enable auto-merge, merge, create a release, or start indefinite monitoring. If explicitly asked to watch CI, continue with pr-ci-watch or its native equivalent.
Report
Return the PR link or clearly labelled draft text, head/base, publication/draft state and concise validation. State any remaining gap.
Provenance
Adapt t3dotgg/tstack file-pr and its comment-formatting discipline. Keep factual attribution and concise writing while removing mandatory model signatures and emoji.
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.

