User repository on GitHub
My JEV Orchestrator to supply my needs
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Agents & MCP
JEV Orchestrator
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.
- Stars
- 5
- Language
- TypeScript
- License
- MIT
- Reading time
- 3 min
- Published by
- Community
- Author
- hiyoku
- Last verified
- Oct 2026
- Languages
- TypeScript
- Platforms
- CLI
- Capabilities
- Open sourceRequires API key
Highlights
- Intercepts Agent spawns: JEV reads the task text and picks the model, effort and tool profile for each subagent.
- Install inside Claude Code: /plugin marketplace add hiyoku/jev-orc then /plugin install jev-orchestrator@jev-orc.
- Works with a TypeSafe key or an OpenRouter key pointed at https://openrouter.ai/api/v1/systemone.
- Status line shows the last routing, tokens, cost and a counterfactual comparison; the decision log holds one JSON line per decision.
- Modes auto, off and explicit, with a fail-open fallback of sonnet at medium effort capped by maxModel and maxEffort.
Quickstart
bash
tail -n 20 ~/.claude/jev-orchestrator/decisions.jsonl
tail -n 20 ~/.claude/jev-orchestrator/decisions.jsonl | jq -c 'del(.ts,.tool_use_id)'Watch out
README reviewed 2026-10-17; the optional multi-model verifier is off by default and the README says its 0.5 confidence cutoff was almost never reached during validation, so recalibration is planned from log data. No independent benchmarks or pricing beyond TypeSafe's published list prices are given.
Questions about this entry
Short answers drawn from the linked project or page.
It is a Claude Code MOD (plugin) that sends every Agent spawn through JEV, TypeSafe's decision model, so JEV can pick the subagent's model (haiku, sonnet or opus), effort and tool profile for you. It also ships plan and run skills and a status line showing routing and cost.
You do not call JEV yourself: once enabled, every Agent spawn of the agents in scope goes through JEV first, which reads the task text and returns the model, effort and tool profile. The tool.call intercepts and applies the effort, and agent.spawn applies model and profile, correlated by tool_use_id.
In a Claude Code session run /plugin marketplace add hiyoku/jev-orc and then /plugin install jev-orchestrator@jev-orc, choose a scope (user, project or local) and add your JEV key. The README says getting started takes two commands and that the plan and run skills come with the plugin.
Yes. Put an OpenRouter key in the typesafeApiKey option and set the JEV endpoint to https://openrouter.ai/api/v1/systemone; the README notes OpenRouter hosts JEV with the same System One API and that the endpoint must be https because the key travels with every call.
The MOD never denies a spawn: if JEV fails (missing key, timeout, network or an HTTP error), the spawn proceeds on sonnet at medium effort, capped by maxModel and maxEffort, and the reason is written to the decision log.
Modes are auto (default: every in-scope spawn routed), off (disabled, status shows jev OFF) and explicit (only spawns whose description starts with the [jev] marker). Toggle with /jev-orc on, off or explicit, or set the mode in the plugin configuration.
The GitHub user hiyoku publishes it under the MIT license; the repository was created 2026-10-08 and pushed 2026-10-13, and it carries an English README plus a Portuguese (Brazil) translation.
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
X: OpenRouter Decision Model Rankings
X: herdr + jev, an agent router
X: A group chat bot with decision room routing
More Jev resources like this
Claude Code plugin that asks Jev a few typed questions before every turn and sets reasoning effort, subagent model and effort, strategy, and the skill a prompt needs; you keep Opus for the conversation.
Jev Agents & MCP#community#catalog-import#logicrw
Community
akramovic1/jev-pilotTypeScript
akramovic1/jev-pilotTypeScriptA 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
romanmeclazcke/codex-siftTypeScriptA Claude Code mod that sends each prompt to Jev through OpenRouter and turns the returned 0-3 score into a reasoning effort level, from low to xhigh. Shadow mode is on by default, so it only logs what it would set until you turn it off.
Jev Integrations & Adapters#community#github#claude-code
Community
gebeer/jev-effortTypeScript
gebeer/jev-effortTypeScriptA Claude Code mod that scores each prompt with TypeSafe Jev and sets Claude's effort level for that turn, mapping four narrow questions, reasoning depth, scope, stakes and ambiguity, onto an effort level without ever touching the model.
Jev Tools & CLIs#community#github#claude-code
A Claude Code mod that routes each request to haiku, sonnet or opus with a reasoning effort and confidence score using local heuristics, and can optionally let Jev (TypeSafe System One) pick the tier instead. Adds a live status band above the prompt.
Jev Tools & CLIs#community#github#claude-code
Community
visualife/ClaudeRouterTypeScript
visualife/ClaudeRouterTypeScriptA Claude Code mod that asks Jev (TypeSafe) how much effort each prompt needs and sends every request of that turn at that effort, never changing the main loop's model. Off by default; a fork of jjjjjjjjjjjjjjjjacob/jev-router.
Jev Tools & CLIs#community#github#claude-code
Community
vntrungld/jeffortTypeScript
vntrungld/jeffortTypeScriptFrom 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.