Skip to content
v1.4.0MIT

Audits a codebase's user-facing copy, layouts and metadata for AI-written tells and factual errors, such as a privacy policy that misses the trackers the code loads, reviews a web, iOS or Android screen for AI-generated design and fixes it in ordered passes once you approve a direction, rewrites a file or pasted text to remove AI writing patterns, prunes code comments that narrate the code or excuse a workaround, prunes tests that break on every refactor, check that a removed feature is gone or can never fail, and rebuilds a feature an agent just wrote as a simpler second version, keeping the first until you choose

slop-tests

Find tests that can't catch a bug (change detectors that break on every refactor, tests that check a removed feature is gone, and tests that can never fail), then plan which to delete, fix or rewrite without losing coverage of real behavior. Can confirm each finding by breaking the code in a scratch worktree. Report-only until the user approves, unless run with --fix. Use when the user runs /slop-tests, or asks to prune, clean up or remove low-value, useless or AI-generated tests.

Read SKILL.md at the source

Pinned to revision 511d47f1a323, so it is the text this page describes rather than whatever the author pushed since.

Pre-approved tools experimental

Experimental field. Support varies between clients, so this list is what the author declared, not what your client will enforce.

  • Read
  • Glob
  • Grep
  • Agent
  • Bash(git diff *)
  • Bash(git log *)
  • Bash(git blame *)
  • Bash(git ls-files *)
  • Bash(git merge-base *)
  • Bash(git symbolic-ref *)
  • Bash(git worktree list *)
  • Bash(git worktree add *)
  • Bash(git worktree remove *)
  • Bash(git stash create*)
  • Bash(git check-ignore *)

Files

Every link opens the file at its source, pinned to the revision this page describes.