agentleFS
dvy1987dvy1987/agent-loom/.agents/skills/pr-authoring/SKILL.md

pr-authoring

Skill4 starsChanged 3 months agoMITView on GitHub

Author pull requests with intent-separated commits, blast-radius disclosure, and verification evidence. Load when opening a PR, writing a PR description, separating refactor from feature commits, or summarizing a change set for review. Also triggers on "write the PR", "PR body", "intent-separated commits", "split refactor from feature", "pr summary", or after safe-change cycles complete. Extends git-workflow-and-versioning — does not replace it.

  • Commits and pushes

4 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 1 is distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Never mix feat/fix commits with refactor/chore formatting in one PR when avoidable.

    line 27
  2. Very specific:

    Every PR body includes blast radius (from dependency-mapping if available) and verification evidence (from safe-change or CI).

    line 28
  3. Very specific:

    Follow commit conventions from git-workflow-and-versioning.

    line 30
  4. Very specific:

    Never omit behaviorVerified: false when tests were not run.

    line 31
  5. Distinctive:

    If mixed, split into separate commits/PRs before proceeding.

    line 47
  6. Very specific:

    Use template in references/PR-CONVENTIONS.md.

    line 55

“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “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

PR Authoring

You produce reviewable PRs: one intent per commit, body that states what changed, why, blast radius, and verification result. Mixed refactor+feature PRs raise review burden — separate them.

Hard Rules

Never mix feat/fix commits with refactor/chore formatting in one PR when avoidable. Every PR body includes blast radius (from dependency-mapping if available) and verification evidence (from safe-change or CI). One primary intent per PR — split if the diff serves two goals. Follow commit conventions from git-workflow-and-versioning. Never omit behaviorVerified: false when tests were not run.


Workflow

Step 1 — Classify commits

Group staged/historical commits by intent:

IntentCommit types
Behavior changefeat, fix
Structure onlyrefactor
Hygienechore, docs, test

If mixed, split into separate commits/PRs before proceeding.

Step 2 — Gather evidence

Collect: impact report (dependency-mapping), verify JSON (safe-change), or CI links.

Step 3 — Write PR body

Use template in references/PR-CONVENTIONS.md. Fill every section — no placeholders.

Step 4 — Pre-flight review

  • Title matches primary intent (feat: / fix: / refactor:)
  • Body lists blast radius
  • Verification section has commands + results
  • No secrets in diff or body

Step 5 — Output for user

Emit the PR title, body (markdown), and suggested gh pr create command if applicable.


Gotchas

  • "While I'm here" refactors belong in a separate PR — reviewers cannot bisect mixed intent.
  • Large PRs without blast radius get rejected — the table is mandatory.
  • Squashing intent-separated commits destroys narrative — prefer merge with history.

Output Format

## PR draft — [title]

**Intent:** [feat|fix|refactor|chore] — [one line]

### Body
[paste PR-CONVENTIONS template filled]

### Commits (intent-separated)
1. `type: message`
2. ...

### Create
gh pr create --title "..." --body "$(cat <<'EOF'
...
EOF
)"

Examples

Teaser: Two commits — refactor: extract parser then feat: add validation — PR body lists 3 affected modules and npm test pass.

Full pairs: references/examples.md


Common Rationalizations

ExcuseReality
"Reviewers can figure out intent"Mixed PRs take 2× review time (empirical).
"One PR is faster"Faster to merge wrong.
"Blast radius is obvious"Write it — forces you to check.
"CI will catch it"CI doesn't explain why you changed it.
"Docs don't need verify section"Docs PRs still need scope + risk statement.

Verification

  • Intent-separated commits listed
  • PR body uses PR-CONVENTIONS template
  • Blast radius section present
  • Verification evidence attached

Red Flags

  • feat + refactor in single commit
  • PR body missing verification section
  • behaviorVerified omitted when no tests ran

Prune Log

Last pruned: 2026-07-05

  • Initial release from high-leverage skill spec (Skill 2 family)

Impact Report

PR drafted: [title] | Intents: N commits | Blast radius: [low|med|high]
Verification cited: [yes|no]

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.