agentleFS
Sign inStart free
rpraharajrpraharaj/ai-sdlc/skills/ready-to-merge/SKILL.md

ready-to-merge

Skill0 starsChanged 19 days agoMITView on GitHub

Decide if the current change is ready to merge. Run this repo's checks; secret-scan; apply engineering-bar; dispatch security, performance, production, 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

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If the diff is empty, stop.

    line 34
  2. Distinctive:

    Commit / open PR (only if they asked)

    line 39

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Ready to merge

Gate the current change. Do not add features. Do not declare ready without this session’s command output.

Is this the right skill?

SignalDo this instead
Still implementingimplement-change
Review comments onlyreview-change
Cut a releasecut-release after this is Ready

Required inputs

  • A non-empty diff
  • AGENTS.md commands
  • context/definition-of-done.md and context/decision-rights.md when installed; identify any missing required check or approval policy before a verdict

If the diff is empty, stop.

Human decisions (do not invent)

  • Waivers (must be explicit)
  • Commit / open PR (only if they asked)

Procedure

  1. Identify the target revision and base. Include staged, unstaged, and untracked changes. For a committed branch with a clean working tree, use its merge-base with the default base; a clean tree does not mean an empty change. Stop if the base cannot be established.
  2. Inventory required checks from AGENTS.md and the program's definition of done. Run format, lint, typecheck, focused tests, Test, and any required build or other check. Record each as pass, fail, not applicable, unknown, or not run, with command, exit status, output, and target revision. not applicable needs a reason; (unknown) is not not applicable. Required checks that are unknown, not run, or fail block Ready unless a human authorized in context/decision-rights.md explicitly waives that check for this revision. No executable check evidence means unverified, not Ready.
  3. Secret scan via engineering-bar. Any match is not ready.
  4. Walk engineering-bar on the diff.
  5. Acceptance: read the shaped slice, named task, and applicable Must requirements. Link each acceptance check to its evidence. Missing acceptance is unverified; use n/a only for an explicitly explained non-behavioral change. Verify required human approvals and waivers using the evidence convention in context/decision-rights.md. A label such as accepted is insufficient. If the diff changes after verification, rerun affected checks and obtain any affected approval again.
  6. Dispatch when the diff matches — load sibling ../NAME/SKILL.md:
Diff signalRun
Auth, input, uploads, HTML, URLs, secrets, PIIsecurity-pass
Query/RPC in a loop, pagination, hot path, large importperformance guard
Network, queue, job, webhook, retry, timeout, flagproduction-pass
Schema, migration, protobuf, OpenAPI, event shapedata-change (check plan; do not start a new migration unless asked)
UI components, routes, CSS, views, client stateverify-ui

Silent skips are not ready.

Output

  • Path: reply
  • Template: references/rtm-template.md

Quality bar

  • Ready only if every required check and dispatch passed or has a revision-bound human waiver, secrets are clean, the bar has no unwaived fails, acceptance is evidenced or explicitly not applicable, and required approvals are current
  • Command output from this session
  • Unknown or unrun required checks cannot silently pass; target revision and base are recorded

Done

The verdict. Next: commit then open-pr if they asked.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free