agentleFS
imMamdouhaboammarimMamdouhaboammar/anti-slop-engineering/skills/pr-boundary-guard/SKILL.md

pr-boundary-guard

Skill4 starsChanged 31 days agoMITView on GitHub

Enforce Git provenance, branch topology, commit message conventions, PR template contracts, and logorrhea bounds. Use when preparing, reviewing, or triaging pull requests in local Git checkouts or automated CI pipelines.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    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. Very specific:

    Topology hygiene: Never submit PRs directly from main or master.

    line 27

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

PR Boundary & Git Topology Guard

Enforce clean Git provenance, branch hygiene, radio-button PR template contracts, and anti-logorrhea bounds across local checkouts and CI pipelines.

Why This Skill Is Needed & Target State

Open-source maintainers and engineering teams face an influx of low-quality, AI-generated PRs characterized by:

  • Direct pushes from unbranched default branches (main/master) on personal forks.
  • Mismatched commit author identities and automated crawler cherry-picks.
  • Excessive metadata logorrhea (4,000-word PR descriptions, 15+ inline code tokens, and emoji inflation).
  • Hallucinated extra sections in PR templates and unselected radio-button sections.
  • Micro-edits (<=3 lines) solely targeting README.md or LICENSE for metric gaming.

The target state is a deterministic, offline-first evaluation pipeline operating on local Git history and CI event payloads, returning machine-readable RFC 9457 problem details and remediation contracts.

Invariant Rules

  1. Topology hygiene: Never submit PRs directly from main or master. Always branch to an intent-revealing feature branch (feat/, fix/, chore/).
  2. Subject length discipline: Limit commit subject lines to 100 characters max (warning at >72 chars). Rationale belongs in the commit body (up to 5,000 chars).
  3. Template radio-button contracts: For mutually exclusive checklist sections, check exactly one option. Do not inject hallucinated sections not present in the repository template.
  4. Emoji and token restraint: Bounded emoji count (<=3) and bounded inline code reference density in PR descriptions. Let clear engineering descriptions speak without marketing fluff.
  5. Shallow and detached resilience: Automatically adapt when running in detached HEAD (refs/pull/*/merge) or shallow clones without crashing or emitting false alarms.

CLI Usage

# Run local git topology and commit check
bunx anti-slop-engineering pr-guard --base origin/main

# Run full PR check with PR description and repository template
bunx anti-slop-engineering pr-guard \
  --base origin/main \
  --pr-title "feat(core): implement robust boundary validation" \
  --pr-body-file .github/pr-description.md \
  --template-file .github/pull_request_template.md \
  --require-conventional \
  --json

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.