Skip to content

Jev directory

Jevlint — ice-hazymoon

A linter whose rules are plain-English yes/no questions about code — secrets in logs, retry loops that run forever, leaked error messages — answered by TypeSafe's Jev as calibrated probabilities, with output shaped like any other linter.
Stars
16
Language
TypeScript
License
MIT
Reading time
2 min
Published by
Community
Author
ice-hazymoon
Last verified
Oct 2026

ice-hazymoon/jevlint

User repository on GitHub

View on GitHub

Semantic lint rules for the code-review questions a deterministic linter can't express

Forks
1
Last pushed
Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.Homepage

This is one of several Jev listings for the same project. The canonical entry is Jevlint — codegirl-007.

Highlights

  • Each rule is one narrow question phrased so that yes means violation; Jev answers with a calibrated probability such as p=0.99 instead of free text.
  • npx jevlint init writes jevlint.config.ts plus an example rule and two fixtures; npx jevlint test proves each rule passes its own fixtures.
  • Prefilters, unless and filePrefilter regexes decide cases in code first, and an optional ast-grep candidates selector gives node-exact report ranges.
  • Files up to 150 lines are judged as one chunk, larger ones are sliced, and a second pass narrows chunks over 40 lines to the window that actually fired.
  • Requires Node.js 22+ or Bun and a TypeSafe or Vercel AI Gateway key; the README says to keep ESLint for anything syntax can decide.

Quickstart

bash
npm install --save-dev @hazymoon/jevlint typescript
npx jevlint init
npx jevlint test
npx jevlint

Watch out

Grounded in the truncated repository README and GitHub metadata reviewed on 2026-10-15; the README publishes no rule-accuracy benchmarks, and it states directly that syntax-decidable questions are cheaper and exact in a deterministic linter.

Questions about this entry

Short answers drawn from the linked project or page.

Reactions & coverage

Posts, threads, and videos about this entry from around the web.

Hacker News: Thread on Selmar's Jev code-review calibration notes and ordering.py comparisons.

Lectem2 pointsCalibrating Jev as a Code ReviewerRead the thread on Hacker News

Hacker News: Jev classifies agent PR hunks as P0, P1, or P2 so reviewers see P0 first.

namanbhulawat47 pointsJev Based Code ReviewRead the thread on Hacker News

X: Magic Jev Ball for code review

X: jev-review: a local-first score loop for coding agents

From the Jev guides

Original write-ups on what this entry is for.

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai