Automated README summary
Automatically summarized from the README on 19 Oct 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
A TypeScript MCP server, CLI runner, and SDK that exposes TypeSafe Jev decision models to coding agents like Claude Code, OMP, Cursor, and Windsurf, plus terminal workflows, for calibrated code review, acceptance gating, and safety screening.
How it uses Jev
Jev models are called through an OpenAI-compatible endpoint or direct decision gateway (JEV_BASE_URL, JEV_MODEL, JEV_API_KEY). Tools like jev_review, jev_action_gate, jev_decide, and jev_judge send state such as git diffs, shell commands, or candidate lists as typed questions and return calibrated probabilities and structured verdicts used for allow/confirm/block, auto decisions, and pass/fail gating.
Primitives:choicescorenoul
Technique worth stealing
Replace chat-model rationales with single-pass typed classification returning calibrated probabilities, cutting latency and token overhead.
Try it
Run `npx -y jev-judge setup` to configure a client, or pipe a diff into `jev review --stdin`.
Model judgments from README excerpts
Project signals use repository metadata, an existing summary and the first 700 README characters, not the full project. “Not found” means not detected in that material; it does not establish absence. These are the original judgments with recorded models and dates, not source audits or run results.
Models & evaluation dates
- Inclusion gate
jev-1.13.0- Project signals
jev-1.13.0- Category
jev-1.13.0- Pattern tags
jev-1.13.0
Evidence
These signals use repository metadata, an existing summary and at most 700 cleaned README characters. ≥ 0.60 means reported in that excerpt; ≤ 0.40 means not found there. The middle band is inconclusive. None is a source-code audit or a test result. The runnable-tool question excludes research experiments; the recommendation question asks for a learning shortlist, not a quality verdict.
- Jev is core to the projectFound in material0.88
- Shows a System One patternunclear0.41
- Describes handling uncertaintynot found in the excerpt0.09
- Reports measured numbersnot found in the excerpt0.06
- Is a runnable toolFound in material0.94
- Top-10 learning shortlist signalnot found in the excerpt0.30
- Is a replica of the modelnot found in the excerpt0.15
- Solves a common problem (0–2)score on a 0–2 scale1.97 / 2
- About JevFound in material0.94
no-judgment band (0.40–0.60)
no-judgment band (0.40–0.60)
The Jev model and date for each judgment stage are listed above. Card written by DeepSeek V4.1 Flash from the README on 19 Oct 2026.