What split of work does the skill create between Jev and Claude?

Jev sorts, checks, scores, gates and verifies; Claude reads the unsure slice and writes drafts, replies and summaries. A rules block in the preset reconciles answers that contradict each other, and a person approves what ships.

Which commands come with the install?

Thirteen: setup, status, classify, check, score, route, verify, gate, match, tune, lint, bench and watch. Routed items act by label when Jev is sure, and only unsure items go to Claude with the full text.

What did Jev cost compared with the Claude model?

On the author's 216-email benchmark Jev cost $0.01 against $4.73 for Claude Opus, roughly 400 times cheaper, with input at about 4 cents per million tokens and free output.

How reliable is the confidence number?

In the benchmark, when Jev said it was at least 70% sure, which covered 82% of the inbox, it matched Claude 95.5% of the time; when it was less sure the agreement dropped to 47%, and it flagged which was which.

Can Jev write my replies for me?

No. It returns a pick, a probability per choice and a confidence, and it has no words to generate with; every draft, reply, summary and explanation still comes from Claude or from you.

What does the /jev-gate command protect?

It answers a yes/no safety question in front of tool calls — approve, block or escalate — and installs as a PreToolUse hook, with each decision sent through the same auditable script.

What is needed to install and run it?

python3 3.8 or newer and an OpenRouter API key; the installer is npx claude-x-jev install --with-commands, and everything runs through one dependency-free file, skill/scripts/jev.py, callable from a terminal too.

CommunityJev Agents & MCP

Claude x Jev

A free Claude Code skill that puts Jev in front of the expensive model: thirteen commands sort, check, score, gate and verify with Jev, so Claude reads only the unsure slice and writes the output. Needs python3 and an OpenRouter key.

See highlights

Open github.com

Stars: 59
Language: Python
License: MIT
Published by: Community
Author: charlesdove977
Last verified: Oct 2026

charlesdove977/claude-x-jev

User repository on GitHub

View on GitHub

Fast, cheap, typed decisions for Claude Code. Jev (TypeSafe's decision model on OpenRouter) sorts, checks, scores, gates and verifies at 0.3s and a fraction of a cent per item. Claude keeps the reading, writing and judgment.

Forks: 8
Last pushed: Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.

Tagscommunity python claude-code routing skill

Works forJev for Classification Jev for Routing Jev for Decision gating

Used inJev for Software & DevTools Jev for Sales & Marketing

Languages: Python
Platforms: CLI
Capabilities: Open sourceRequires API key

Highlights

Installs with npx claude-x-jev install --with-commands; /jev-setup stores one OpenRouter key and makes a live test call.
Thirteen commands cover setup, status, classify, check, score, route, verify, gate, match, tune, lint, bench and watch.
On 216 inbox emails with four questions, Jev cost $0.01 against $4.73 for Claude Opus, at 0.3 s per email.
When Jev was at least 70% sure, covering 82% of the inbox, it matched Claude 95.5% of the time; below that, 47%.
/jev-gate installs as a PreToolUse hook that approves, blocks or escalates a tool call in about 0.3 seconds.

Watch out

MIT, installed with npx, needs python3 3.8+ and an OpenRouter key. Jev cannot write text or connect its answers, so a rules block handles cross-question conflicts, and the README's benchmark numbers come from one inbox of 216 emails rather than a public dataset.

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.

Facebook: Intake Router keeps diagnosis out of the answer space at the front desk.

FacebookClinicnotes

YouTube: Jev explained for routing and judging, with a comparison against LLMs.

Hacker News: Vercel on using Jev for defined decisions and GPT-6 Astra when the job also has to write or use tools.

flashbrew4 pointsTypeSafe AI Jev vs. GPT-6 AstraRead the thread on Hacker News

Hacker News: ASTGL replay chose local Laya routing over hosted Jev on one Mac Studio setup.

Jmeg8r15 pointsDoes an open-weight decision model beat a hosted one? Jev vs. LayaRead the thread on Hacker News

Related terms

Glossary definitions related to this entry.

Intent routing

More Jev resources like this

Gatekeeper

Claude Code hooks that narrow a roster in YAML, ask Jev which handler owns a prompt, and log the band before anything is blocked.

Jev Agents & MCP#community#python#coding-agent

Community

AgriciDaniel/gatekeeperPython

Jev Hermes Router

A Hermes plugin that routes every message you type to one of four model levels with Jev deciding in about 300 ms: lightweight for the trivial, maximum for what matters, and continuations stay on the model with the warm cache.

Jev Agents & MCP#community#python#hermes

Community

okjpg/jev-hermes-routerPython

Jev Harness

An agent harness where Pi runs the loop, Gemini does the work, and Jev makes the small decisions: which model tier to use, which runbook sections to load, whether a tool call may run on a real Postgres database, and whether the report holds up.

Jev Agents & MCP#community#python#agents

Community

PromtEngineer/jev-harnessPython

Codex Sift

A model router for the Codex CLI: on every user turn Jev judges how hard the task is and a YAML policy picks one of three lanes — flash, craft or forge — so premium quota goes to the prompts that need it.

Jev Agents & MCP#community#typescript#codex

Community

romanmeclazcke/codex-siftTypeScript

Jev Router

Launchers for Claude Code, Codex, Grok CLI, and opencode that ask Jev which model-and-effort tier each human turn needs, rewriting only the model field while the real CLI runs unchanged.

Jev Agents & MCP#community#javascript#routing

Community