
A checklist for stripping AI fingerprints from text. It catches the usual suspects: significance inflation, em dash addiction, "delve" and "landscape" and "testament," plus structural tells like inline-header lists and Title Case Everything. The second half matters more than the first. Removing clichés leaves you with bland, correct prose that still sounds like a bot. The fix is adding opinion, varying sentence rhythm, and saying concrete things instead of wrapping simple points in abstract framing. Works as both a post-draft editing pass and a way to train yourself out of these patterns if you're using LLMs to draft. Honestly, if you write with AI assistance at all, you need this running by default.
npx -y skills add cursor/plugins --skill unslop --agent claude-codeInstalls into .claude/skills of the current project.
Edit text to remove AI patterns and add human voice.
Removing patterns is half the job. Sterile, voiceless writing is just as obvious.
.toSQL() returns the exact string sent to the database", "a column rename fails the build". Ask what the sentence tells the reader to do or know, then write that. If you can't restate it as a concrete instruction, fact, or number, cut it. One more check: if the sentence could appear unchanged in another project's docs, it says nothing about this one. Cut it.