LiteLLM Showcases 'anti-ai-slop' Skill to Detect and Remove AI Writing Patterns

LiteLLM (YC W23) ·

Key Info

LiteLLM highlights "anti-ai-slop", a four-stage skill that detects and removes AI-generated writing patterns ("slop") from codex/cc text, catching 41 distinct patterns across ~120 AI words and ~100 phrases.

Highlights

  • Identifies common AI tells: em dashes, emoji bullets, title-case headings, watermarks, negative parallelism, colon reveals, and throat-clearing openers.
  • Four gates: sanitize invisible characters, remove common AI words/chars, have an agent review original vs rewrite with loop-back on failures, and optionally use a parallel agent with a different model for extra review.
  • Aimed at making AI-assisted writing sound more natural and human.
Loading...