agentleFS
WendorWendor/teapod-stream/.claude/skills/pr/SKILL.md

pr

Skill558 starsChanged todayMITView on GitHub

Use when opening or updating a pull request in teapodstream — runs the CI checks locally, fills the PR template honestly and keeps the description current.

  • Commits and pushes

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is 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.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run every check and read the output — don't assume:

    line 16
  2. Very specific:

    No placeholders left (<…>): the PR rules workflow fails on them.

    line 35
  3. Distinctive:

    Never apply a suggestion you haven't verified.

    line 60

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Opening or updating a pull request

Rules: CONTRIBUTING.md. Template: .github/pull_request_template.md.

Before opening

  1. git diff master...HEAD --stat: one topic? Unrelated hunks → move them to another branch. Over ~500 lines (without lib/l10n/) → split, or the issue must already agree on the size.
  2. Nothing from the hands-off list in AGENTS.md changed (version, build.sh, .github/, signing, android/app/libs/) unless that is the task.
  3. Run every check and read the output — don't assume:
    dart format --output=none --set-exit-if-changed lib test tool
    flutter analyze && flutter test
    (cd android && ./gradlew :app:testDebugUnitTest)
  4. Wiki pages for the touched area still true? Fix them in this branch.
  5. Grep the diff for secrets: git diff master...HEAD | grep -nE 'vless://|vmess://|trojan://|ss://|hy2://|[0-9a-f]{8}-[0-9a-f]{4}-'.

The description

  • Why: the user-visible problem, one or two sentences; Fixes #N / See #N.
  • What changed: summary, then - area: change (file, function) for what a reviewer should read first.
  • Testing: list only what was actually run. You can't run a device: write native/VPN behaviour under Not tested and ask the person you work for to run it — never claim a device test you didn't see.
  • Risks: connection, kill switch, reconnect, stored data migration; "None" only after checking.
  • Tick a box only when it is true. Keep the AI was used box.
  • End the description and every comment or review reply you post with Co-Authored-By: <agent> <email>; commits you make carry it as a trailer.
  • No placeholders left (<…>): the PR rules workflow fails on them.

Open and update

gh pr create --draft --base master --title "area: what changed" --body-file <filled template>
gh pr ready <n>            # when checks pass and the description is complete
gh pr edit <n> --body-file <updated description>   # after every push that changes what the PR does

The description states the PR as it is now; review history stays in the comments.

CodeRabbit

It reviews when the PR leaves draft and again after each push.

gh pr checks <n> --watch                       # wait until the CodeRabbit check finishes
gh api repos/Wendor/teapod-stream/pulls/<n>/comments --jq '.[] | select(.user.login|test("coderabbit")) | {id, path, line, body}'
gh api repos/Wendor/teapod-stream/pulls/<n>/reviews --jq '.[] | select(.user.login|test("coderabbit")) | .body'
                                               # every review body: outside-diff and nitpick items live only here

For each comment: check it against the code and the wiki. Valid → fix in a new commit. Wrong or out of scope → reply on the thread with the reason (gh api repos/Wendor/teapod-stream/pulls/<n>/comments/<id>/replies -f body=...). Never apply a suggestion you haven't verified; never leave one unanswered. Repeat after its re-review.

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.