The "no-ai-slop" project, developed by petergyang, is a specialized skill designed to identify and remove over 20 distinct linguistic patterns characteristic of generative AI output while preserving an author's unique voice. The tool targets specific stylistic tropes such as binary contrasts (e.g., "It’s not X. It’s Y."), throat-clearing openers, faux-insight setups, colon reveals, and dramatic fragments. It also mitigates "importance puffery" and "weasel attribution" to prevent superficial or vague writing. Beyond pattern recognition, the skill enforces fundamental writing improvements by prioritizing active voice, leading with the main point, untangling complex sentence structures, and substituting abstractions with concrete details. Users can implement the tool via a global installation using npx (npx skills add petergyang/no-ai-slop --skill no-ai-slop --global) or by integrating it directly into LLM environments like ChatGPT, Claude Code, or Codex. The repository includes a structured workflow defined in SKILL.md, an evaluation framework in eval.md, and a plugin build system via build_plugin.py. When active, the tool not only scrubs identified "slop" patterns but also provides transparency by quoting every specific instance it detects, allowing users to audit the edits made to their original text.