Skip to content

Latest commit

 

History

History
22 lines (15 loc) · 1.15 KB

File metadata and controls

22 lines (15 loc) · 1.15 KB

Credits & prior art

Slop Cop didn't invent the idea of hunting AI slop. It synthesizes and de-conflicts several open-source projects and a body of recent research into one routed skill with a forced scoring procedure.

Source projects

  • stop-slop — Hardik Pandya
  • slopless — seochecks.ai
  • hallmark — Nutlope
  • taste-skill — Leonxlnx
  • desloppify — Peter O'Malley
  • Rossmann Group no-AI-slop writing rules

Research the beats were hardened with

  • Prose — updated for the 2026 landscape, where the em dash is a fading tell and cadence uniformity is the strongest signal.
  • Design — generated-UI research, including the roughly 50ms credibility judgment and the indigo-gradient / Inter default cluster.
  • Code — LLM code-smell findings: happy-path scope, over-documentation of the obvious, and higher smell density on more complex tasks.

What Slop Cop adds

A single routed entry point across prose, design, and code; a false-positive-prevention layer so it tickets slop and not vocabulary; and a forced scoring procedure with calibration anchors, so grades stay consistent instead of drifting on gut feel.