What it is
A small TypeScript code-review workflow for developers: it reviews a Git diff or scans a whole codebase and shows results in a quiet local dashboard bound to 127.0.0.1.
How it uses Jev
Jev makes bounded judgments in a staged pipeline: a Noul risk matrix, Choice and Score file profiles, Choice evidence selection, Choice mechanism classification, Score severity, and conditional Choice reviewer routing. Orchestration, thresholds, and workflow policy stay in code; Jev returns typed judgments the workflow uses to advance each stage.
Primitives:choicescorenoul
Technique worth stealing
Keep orchestration in code and use Jev only for bounded, typed judgments at each stage.
Try it
npm install; add TYPESAFE_API_KEY to .env; npm run review:changes:save -- /path/to/repo; npm run dashboard.
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-centricyes0.92
- Shows a System One patternyes0.83
- Handles uncertaintyno0.07
- Measuredno0.04
- Runnableyes0.76
- Worth recommendingunclear0.58
- Model replicano0.05
- Problem scopescore on a 0–2 scale1.93
- About Jevyes0.98
Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 20 Sept 2026.