agentleFS
sebastienrousseausebastienrousseau/agtmls/skills/anti-slop-pr-and-writing/SKILL.md

anti-slop-pr-and-writing

Skill1 starsChanged 3 months agoApache-2.0View on GitHub

Eliminate AI slop, clichés, sycophancy, and robotic filler from PR descriptions, commit messages, code comments, and technical documentation. Load when reviewing or authoring PRs, drafting release notes, or stripping conversational apologies and binary contrasts while preserving human voice and technical facts.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are 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.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Preserve technical density: Never replace a specific error code, function name, or metric with vague prose.

    line 55
  2. Distinctive:

    Keep the author's cadence: Do not normalize unique idioms into generic corporate-speak.

    line 56
  3. Distinctive:

    Wrap strictly: Commit bodies at 72 characters, headers at 50 characters, following Conventional Commits.

    line 57

≈ “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

Anti-slop PR and writing

Trigger. You are writing or reviewing Pull Request descriptions, git commit messages, architecture documentation, or release notes. Also load whenever text has accumulated conversational filler, robotic clichés, superficial summaries, or sycophantic phrasing.

Reference material (the 20+ specific slop catalog, before-and-after tables, and diff cleaning patterns): see reference.md.

The core doctrine

Engineers read diffs to understand intent and mechanics. AI slop obscures intent, wastes review attention, and destroys credibility.

Good technical writing is dense, precise, and human. When editing with or through an agent, eliminate artificial mannerisms without flattening the author's voice or stripping technical context.

Five patterns to strip on sight

1. Conversational sycophancy and robotic apologies

  • Strip: "Certainly!", "I would be delighted to help with that!", "As an AI language model...", "I apologize for the oversight."
  • Replace with: Direct statements of action, findings, or code changes.

2. Throat-clearing openers

  • Strip: Temporal clichés about fast-paced eras, "Here's the thing...", "Let's dive in...", "It goes without saying that..."
  • Replace with: The actual subject, problem, or decision in sentence one.

3. Binary contrasts and fake profundity

  • Strip: "It's not X. It's Y.", "The future isn't coming; it's already here.", "This changes everything."
  • Replace with: Concrete technical trade-offs and empirical measurements.

4. Fluffy PR summaries and emoji theater

  • Strip: Rocket emojis (🚀), party poppers (🎉), and bulleted lists that merely re-state trivial git diff filenames.
  • Replace with: The Why (the underlying problem or issue) and the What (the architectural choice made).

5. Defensive and trivial code comments

  • Strip: Comments that repeat obvious syntax (e.g. // increment i, // set status to true, // handle error).
  • Replace with: Comments explaining invariants and non-obvious rationale, or no comment if the code is self-documenting.

The Human-Voice Preservation Checklist

When cleaning text:

  1. Preserve technical density: Never replace a specific error code, function name, or metric with vague prose.
  2. Keep the author's cadence: Do not normalize unique idioms into generic corporate-speak.
  3. Wrap strictly: Commit bodies at 72 characters, headers at 50 characters, following Conventional Commits.

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.