What it is
jev-acp is a standalone ACP agent that evaluates explicit Choice, Score, and Noul questions and returns typed answers, probabilities, and confidence without an intermediate LLM. It is for ACP clients and IDEs.
How it uses Jev
Jev evaluates user-supplied options, rubrics, or statements. The agent collects input through guided tasks, sends it to Jev, and displays the selected option, confidence, and probability distribution. Results are returned as typed answers.
Primitives:choicescorenoul
Technique worth stealing
Guided task picker with reusable templates and explicit review before evaluation.
Try it
npx -y jev-acp
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-centricyes0.97
- Shows a System One patternno0.38
- Handles uncertaintyno0.05
- Measuredno0.04
- Runnableyes0.95
- Worth recommendingunclear0.41
- Model replicano0.05
- Problem scopescore on a 0–2 scale1.00
- About Jevyes0.98
Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 21 Sept 2026.