Jev in production › SDKs and workflow integrations

use-jev

use-jev is a Claude Code and Codex skill that hands off bounded semantic decisions to Jev via OpenRouter, returning a typed choice or score instead of prose.

Open on GitHub ↗

The source publishes no measured number.
Use
SDKs and workflow integrations
Industry
Developer tools
Form
Open-source tool
Stage
Beta
Plugs into
Claude Code, Codex
Listed
2026-09-30
Found via
github
Repository
Kmasterrr/use-jev
Stars
2
Forks
1
Last push
2026-09-29
Language
JavaScript
License
MIT
use-jev screenshot
docs/overview.webp in the Kmasterrr/use-jev README, MIT; shown from GitHub.

The README opens with

A skill that lets Claude Code and Codex hand off bounded semantic decisions to Jev (typesafe/jev-1.13), TypeSafe's System One decision model, through OpenRouter.

You give Jev some text and a typed question. It gives you back a choice, a score, or a probability — with calibrated confidence — instead of prose. Your agent stays in charge of the workflow and writes anything a human will read.

A general-purpose model asked "which queue does this ticket belong in?" will think out loud, hedge, and hand you a paragraph you then have to parse. That is slow, costs more than the decision is worth, and gives you no usable measure of how sure it was.

Badge

For the project's own README, linking back here:

Listed in Jev in production

Also used for sdks and workflow integrations