jevbooks

← All projects

jev-spec

nozomi-koborinai/jev-spec · Homepage

Semantic verification engine for AI code and specs, powered by Jev.

Verification & Guardrails88%Runner-up: Evaluation & Benchmarking

What it is

jev-spec is a Specification-Driven Development verification tool for teams using AI coding assistants. It checks whether code satisfies written requirements, catching semantic gaps that structural linters miss.

How it uses Jev

Jev evaluates typed decision primitives over shared specification and implementation context in a single forward pass, returning calibrated probabilities. jev-spec asserts these against thresholds like minProbability, maxProbability, allowedChoices, and minScore, producing pass/fail exit codes for CI and git hooks.

Primitives:choicescorenoul

Technique worth stealing

Use calibrated probability thresholds as deterministic semantic assertions in CI instead of free-form LLM review.

Try it

bunx jev-spec check (or npx jev-spec check); use --mock for offline runs.

View on GitHub

judged by Jevjev-1.13.0

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-centricunclear0.55
  • Shows a System One patternno0.09
  • Handles uncertaintyno0.06
  • Measuredno0.05
  • Runnableyes0.89
  • Worth recommendingno0.23
  • Model replicano0.11
  • Problem scopescore on a 0–2 scale1.55
  • About Jevyes0.94

Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 21 Sept 2026.