What it is
slop-grader is a TypeScript CLI that grades text files against custom or built-in rulesets for AI slop, grammar, and technical doc quality, then helps an AI agent auto-fix violations.
How it uses Jev
Jev answers plain-text questions about each line or the whole document, defined as Line Rules and Document Rules in Markdown. The returned scores and line-level flags are used to produce a human-readable or JSON report that an AI agent then turns into fixes.
Primitives:choicescore
Technique worth stealing
Express any plain-text question about a single line or whole document as a Markdown rule with true/false criteria, then let Jev score it.
Try it
npx @lukstei/slop-grader@latest -r no-ai-slop examples/slop.md
Evidence
Each line is one question put to Jev about the README. ≥ 0.60 reads as yes, ≤ 0.40 as no; in between Jev is not making a call.
- Jev-centricyes0.81
- Shows a System One patternno0.33
- Handles uncertaintyno0.05
- Measuredno0.06
- Runnableyes0.97
- Worth recommendingno0.31
- Model replicano0.06
- Problem scopescore on a 0–2 scale1.25
- About Jevyes0.95
Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 21 Sept 2026.