agentleFS
refpathrefpath/chatgpt-team/.agents/skills/worktree-integration/SKILL.md

worktree-integration

Skill0 starsChanged 2 months agoMITView on GitHub

Assemble reviewed work in a dedicated integration worktree using exact SHAs, then run final repository validation without silently redesigning authored changes.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is 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.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create a dedicated integration branch/worktree from the intended base.

    line 22
  2. Distinctive:

    If user-visible behavior changed, run the integrated UI-validation gate.

    line 27
  3. Distinctive:

    Never substitute an unreviewed moving branch tip for a reviewed SHA.

    line 31
  4. Very specific:

    Use package-readiness for assigned package candidates and technical-writing with unslop for PR, release, and migration prose.

    line 41

≈ “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

Worktree integration

Inputs

For each candidate, require:

  • task ID;
  • base SHA;
  • reviewed head SHA;
  • complete reviewed base-to-head range or ordered commit list, with prerequisites;
  • task attempt/state and relevant contract/dependency revision;
  • validation summary;
  • known risks and omissions.

Procedure

  1. Create a dedicated integration branch/worktree from the intended base. Reuse one only after inspecting its state and preserving retained work.
  2. Integrate exact reviewed SHAs in dependency order.
  3. Resolve only mechanical conflicts where intent is unambiguous.
  4. Return semantic conflicts or changed assumptions to the owning implementation agent/Lead.
  5. Build the assembled application and run repository-required integration checks.
  6. If user-visible behavior changed, run the integrated UI-validation gate.
  7. Record the final integration SHA and evidence.
  8. Push/open a PR only when repository policy permits.

Never substitute an unreviewed moving branch tip for a reviewed SHA.

Require affirmative completed review evidence for that candidate. Neutral, skipped, budget-exhausted, interrupted, or stale checks do not satisfy this gate, even if the hosting provider accepts their status. Return incomplete gates to the coordinator without starting a review loop yourself.

Confirm what the review covers before choosing an integration operation. A named head commit is not automatically the whole change set for cherry-picking. Record the source-to-integrated mapping and verify that the assembled tree includes the complete approved change. Keep approval bound to the candidate and attempt it actually reviewed.

If a writer hands off a patch because Git writes were denied, an authorized coordinator must record and validate its commit before it enters this reviewed-candidate workflow. For QA evidence, distinguish product SHA, authored test SHA, and combined revision exercised. Record relevant dirty state with command results; a clean-looking HEAD alone cannot identify uncommitted tested code.

Integrate assigned documentation changes before final checks. Run affected docs/example checks and inspect relevant generated pages. New edits or conflict resolutions that alter reviewed content require a new candidate revision and refreshed affected review and validation. Ordinary assembly still requires integration evidence for the combined result.

Use package-readiness for assigned package candidates and technical-writing with unslop for PR, release, and migration prose. Retain failing QA reproducers until the defect is resolved; do not remove evidence solely to make the branch green.

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.