agentleFS
gleicongleicon/devskills/skills/ds-deslop/SKILL.md

ds-deslop

Skill20 starsChanged 46 days agoMITView on GitHub

Remove AI-generated code slop introduced on the current branch and align it with the surrounding codebase.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    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:

    When invoked, diff the current branch against the main branch and remove the slop the branch introduced.

    line 7
  2. Distinctive:

    Treat any argument as scope (files or directories).

    line 13
  3. Distinctive:

    Keep behavior unchanged unless fixing a clear bug.

    line 27
  4. Distinctive:

    Prefer minimal, focused edits over broad rewrites.

    line 28

≈ “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

When invoked, diff the current branch against the main branch and remove the slop the branch introduced. Preserve behavior.

The test for slop is simple: code an experienced engineer working in this language and this codebase wouldn't have written. Judge against the surrounding code and the language's idioms — not a fixed checklist.

Arguments

Treat any argument as scope (files or directories). With no scope, review the branch's diff against the main branch.

What to look for

  • Narrating comments — comments that restate what the code does, or break the file's existing comment density and style.
  • Defensive overkill — error handling or guards abnormal for a trusted path. The form is language-specific: a swallowed try/catch, an if err != nil that can't fire, an unwrap_or masking a real invariant, a null check on a value that's never null.
  • Type escape hatches used only to dodge the checker — judged by the language's norms: any / as / @ts-ignore in TypeScript or # type: ignore in Python are usually slop; any / interface{} in Go is often idiomatic. Flag the ones that exist only to silence a type error, not the idiomatic uses.
  • Needless nesting — deep conditionals that read better as early returns or guard clauses (or the language's equivalent: ? in Rust, match, and so on).
  • Speculative scaffolding — unused helpers, premature abstractions, config knobs with one possible value, placeholder TODOs.
  • Anything else inconsistent with the file and surrounding code.

Guardrails

  • Apply the principle, not the syntax — what's slop in one language is idiomatic in another.
  • Keep behavior unchanged unless fixing a clear bug.
  • Prefer minimal, focused edits over broad rewrites. Match each file's existing conventions; do not impose a new style.

Output

Apply the edits, then give a concise 1–3 sentence summary of what was removed or simplified.

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.