Jev directory
CommunityJev Agents & MCP
ClauJev
Claude Code plugin that puts Jev behind six hooks: request triage, automatic effort, a destructive-command guard, rule review, and a final completion check. One file, zero dependencies, MIT licensed.
- Stars
- 4
- Reading time
- 2 min
- Published by
- Community
- Added
- Oct 2026
Highlights
- Listens to six Claude Code hooks and asks Jev typed questions about the request, command, or diff
- About 400 ms per check, and read-only commands skip Jev entirely
- Six days of daily use cost the author US$ 0.07
- Answers land in low, uncertain, or high bands; only high acts, and only in enforce mode
- Never deletes or reverts anything; any error is logged and Claude carries on
Questions about this entry
Short answers drawn from the linked project or page.
Run /plugin install claujev --marketplace Civalski/claujev inside Claude Code, enter your TypeSafe AI key, then restart. An optional mod adds the status line and automatic effort. A single session can be run without it via CLAUJEV_DISABLED=1 claude.
Yes. The key is the only thing the install asks for, and it is stored in your system's secure storage. The project is a community effort and is not affiliated with TypeSafe AI or Anthropic.
Each check takes about 400 ms, and read-only commands skip Jev entirely. Checks run against small states built per hook: the request, the command, or the diff, rather than the whole conversation.
It sends the request, commands, edited snippets, the turn's diff and the commands Claude ran, after removing keys and tokens. Logs stay in ~/.claude/claujev/ and hold answers plus a content hash, never the content.
Actions are reversible by design and thresholds are high on purpose; the uncertain band never acts. Only high-confidence answers act, and only in enforce mode. A false alarm costs one sentence from Claude.
It is MIT licensed, free to use, modify and redistribute including commercially, as long as the copyright notice is kept. It was created by Alisson Civalski as a community project.
Jev triage maps the request to low, medium, high, or xhigh effort. The mod's observe mode shows the suggestion in the status line; on applies it within your limits. Max is never picked automatically and explicit effort requests are not overridden.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: A usage router for Grok Bot
X: A prompt box that fills itself in
X: Vercel's fx safety reviewer, measured at 18× faster
X: How Jev makes agents faster and cheaper
More Jev resources like this
Claude Code and Codex skill that audits a codebase for LLM call sites, proposes which routing, classification, and rubric-scoring decisions to convert to Jev, and reports latency and cost comparisons after the scope is approved.
Jev Agents & MCP#github#claude-code#codex
CommunityKevin-Zhouu/jevify
A Claude Code plugin that sends every Agent spawn through JEV (TypeSafe), which picks the model (haiku, sonnet, opus), the effort and the tool profile of each subagent. It fails open to sonnet at medium effort, accepts a TypeSafe or OpenRouter key, and ships plan and run skills.
Jev Agents & MCP#community#github#claude-code
Community
hiyoku/jev-orcTypeScript
hiyoku/jev-orcTypeScriptZero-dependency supervisor extension for the Pi coding agent. Blocks destructive commands and hardcoded secrets with a 0 ms regex gate, warns on unverified edits, and routes diffs of 10 lines or more through Jev, OpenRouter, or Cloudflare Clef.
Jev Agents & MCP#coding-agent#guardrails#routing
Community
An experimental coding agent CLI built on opencode whose v2.0.0 architecture adds a JEV decision layer with five decision gates, shadow mode, and ECE calibration, plus a three-level RSI self-evolution loop and six Pi-framework extensions.
Jev Agents & MCP#community#github#typescript
Community
momozi1996/momo-codeTypeScript
momozi1996/momo-codeTypeScriptReusable project context and specialist workflows for the AI tools you already use: agents, skills, rules, and MCP tools for Claude Code, Codex, Copilot, and Cursor, plus Jev decision support that offloads bounded LLM decisions to save tokens.
Jev Agents & MCP#community#github#agents
Community
jmagly/aiwgTypeScript
jmagly/aiwgTypeScriptA 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.
Jev Agents & MCP#community#python#claude-code
Community
charlesdove977/claude-x-jevPython
charlesdove977/claude-x-jevPythonFrom 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.