ready-to-merge
Skill1 starsChanged 18 days agoMITView on GitHub
Decide if the current change is ready to merge. Run this repo's format, lint, typecheck, and tests; secret-scan the diff; apply engineering-bar; dispatch security-pass, performance guard, production-pass, data-change, or verify-ui when the diff matches. Use when the user says ready to merge, are we done, definition of done, or ship check.
2 other projects use a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- rpraharaj/ai-sdlcedited, first seen Nov 1, 2026
- flurdy/agent-skillstheir own version, first seen Aug 9, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 1
- rpraharaj/ai-sdlc★ 016% same
Wrote their own version · 1
- flurdy/agent-skills★ 9same 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 repository34 other files it gives its agents.
- AGENTS.mdAGENTS.md
- templates/AGENTS.mdAGENTS.md
- commitSkill
- cut-releaseSkill
- data-changeSkill
- debugSkill
- engineering-barSkill
- explain-codeSkill
- explore-codebaseSkill
- fix-ciSkill
- handoffSkill
- implement-changeSkill
- incident-triageSkill
- lean-reviewSkill
- open-prSkill
- performanceSkill
- production-passSkill
- record-decisionSkill
- refactor-safelySkill
- resolve-conflictsSkill
- revert-changeSkill
- review-changeSkill
- security-passSkill
- shape-workSkill
- ship-notesSkill
- start-projectSkill
- upgrade-depsSkill
- verify-uiSkill
- write-architectureSkill
- write-designSkill
- write-dev-tasksSkill
- write-docsSkill
- write-prdSkill
- write-testsSkill
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 7 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Read AGENTS.md if present.
- Distinctive: line 20
If the diff is empty, say so and stop.
- Distinctive: line 28
Format: check (avoid unrelated rewrites)
- Distinctive: line 32
Build: when compilation, packaging, or generated artifacts can be affected
- Distinctive: line 46
Use the agreed task and carry forward its relevant upstream requirements.
- Distinctive: line 73
All applicable configured checks passed or have an explicit user waiver
- Distinctive: line 77
Every dispatched pass reported pass or an explicit waiver
- Very specific: line 93
Do not commit or open a PR unless the user asked.
≈ “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
Ready to merge
Gate the current change. Do not add features. Do not declare ready without this session's command output.
Read AGENTS.md if present.
Collect the diff
Load engineering-bar and follow its Shared procedures for change scope and verification evidence. Include branch commits, staged and unstaged edits, and relevant untracked files. Pass the recorded scope to every dispatched check.
If the diff is empty, say so and stop.
Always run
1. Mechanical checks
Discover commands from AGENTS.md, manifests, and CI. Apply the shared evidence procedure, including reuse of unchanged evidence; do not silently skip unknown or unexecuted checks:
- Format check (avoid unrelated rewrites)
- Lint
- Typecheck
- Focused tests for the changed area, then Test
- Build when compilation, packaging, or generated artifacts can be affected
Record each command and exit status or the reason it could not run. Use the shared completion rule to decide whether equivalent evidence suffices or the result remains unverified. Fix failures only when authorized; otherwise report them.
2. Secret scan
Load engineering-bar (paths in that skill). Apply Secret files. Any match is not ready.
3. Engineering bar
Walk the rest of engineering-bar on the diff. Fails must be fixed or explicitly waived by the user.
4. Acceptance checks
Use the agreed task and carry forward its relevant upstream requirements. Do not let an older shape override a newer user instruction. Sources, from the most specific to broader context:
shape-workacceptance checks- The named
write-dev-taskstask (Acceptance + Batch done if this change claims the whole batch) write-prdSuccess checks (only those this change was meant to satisfy)
Mark each check pass/fail. Any fail is not ready. If none of those specs exist, write n/a.
Dispatch (only when the diff matches)
Load each sibling from ../NAME/SKILL.md, else .agents/skills/NAME/SKILL.md, else ~/.agents/skills/NAME/SKILL.md. Follow that skill on this diff. If the skill is missing, run the one-line fallback and say it was a fallback.
| Diff signal | Run | Fallback if skill missing |
| Auth, user input, uploads, HTML, URLs, secrets, PII | security-pass | New reachable surface must be authorized; treat input as untrusted; no secrets in the diff |
| Query or RPC in a loop, pagination, hot path, large bundle/import, "slow" | performance in guard mode | No new N+1, no unbounded list/job, no huge eager import on a hot route |
| Network, queue, job, webhook, retry, timeout, feature flag | production-pass | Timeouts on I/O; retries bounded and idempotent; one rollback sentence |
| Schema, migration, protobuf, OpenAPI, event shape, column drop/rename | data-change (check the plan; do not start a new migration unless the user asked) | Expand and contract are not in the same change |
| UI components, routes, CSS, views, client state | verify-ui | New controls have a name, are keyboard-reachable, and are not unreadable; exercise the changed flow, not only a screenshot |
Mark irrelevant passes not applicable with a reason. Only user-approved gaps are waivers. Silent skips are not ready. Run data-change in assessment mode without creating or applying migrations.
Verdict
Ready only if:
- Scope is complete and verification meets the shared completion rule
- All applicable configured checks passed or have an explicit user waiver
- Secret scan clean
- Engineering-bar has no unwaived fails
- Acceptance checks (if any) pass
- Every dispatched pass reported pass or an explicit waiver
Otherwise not ready (failed requirement) or unverified (insufficient evidence), with the blocking item first. Never convert missing evidence into a pass.
Done
## Ready to merge: yes | no | unverified
- Mechanical: <command + exit status, one per line>
- Secrets: pass | fail
- Engineering bar: pass | fails… | waivers…
- Acceptance: pass | fail | n/a
- Dispatched: <skill → pass/fail/waived/skipped and why>Do not commit or open a PR unless the user asked. Next: commit then open-pr.
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.