Jev Flash Router — AI skill for Claude Code
open-sourced jev-flash-router: an MCP server for TypeSafe's new Jev model.
How to install Jev Flash Router
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open Ravinder82/jev-flash-router and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Jev Flash Router does
open-sourced jev-flash-router: an MCP server for TypeSafe's new Jev model. AI coding agents waste hundreds of reasoning tokens just deciding which file to edit, which route to pick, or whether a diff breaks tests. Jev evaluates state and outputs calibrated probabilities. Works with Cursor, Windsurf, & Claude Code
Alternatives in AI
- System Prompts Leaks — Extracted system prompts from ChatGPT (GPT-5.4, GPT-5.3, Codex), Claude (Opus 4.6, Sonnet 4.6, Claude Code), G 38.6k ★
- Optimization — ai-research-skills AWQ, GPTQ, GGUF, Flash Attention, bitsandbytes 5.4k ★
- LLM For Zotero — An open-sourced research agent system deeply rooted in your Zotero library 2.8k ★
README
jev-flash-router
[](https://www.npmjs.com/package/jev-flash-router) [](LICENSE) [](https://nodejs.org) [](https://modelcontextprotocol.io) [](https://openrouter.ai)
**Zero-token-output decision router MCP server powered by TypeSafe Jev.**
AI coding agents waste hundreds of reasoning tokens just deciding which file to edit, which route to pick, or whether a diff breaks tests. **jev-flash-router** evaluates context, code diffs, logs, or planning options and returns calibrated probabilities in ~150ms — with **$0.00 output token cost**.
✨ Features
- ⚡ ~150ms decisions — Lightning-fast evaluation via OpenRouter Decisions API
- 💸 Zero output token cost —
output_tokens: 0on every call - 🧠 TypeSafe Jev — Powered by the cutting-edge
~typesafe/jev-latestmodel - 🔌 MCP-native — Works seamlessly with Cursor, Windsurf, Claude Desktop & Claude Code
- 📦 Zero-config — Install via
npxand go - 🎯 Three decision formats:
noul(binary),choice(categorical),score(rubric)
🚀 Quick Start
1. Install as an MCP server
No local clone needed — anyone can run it instantly:
npx -y jev-flash-router
Or install globally via npm:
npm install -g jev-flash-router
2. Set your OpenRouter API key
Set the `OPENROUTER_API_KEY` environment variable with your [OpenRouter](https://openrouter.ai) API key (required).
3. Use the tool
In your IDE's agent chat, call:
**evaluate_decision**
state: The context, code diff, error log, or task descriptionquestion: The targeted question to evaluatetype:"noul","choice", or `"score"
Related Skills
Jev Router
Route to the cheapest model in claude code for your task using jev-router
Skillranker
Rust CLI powered by Jev from TypeSafe.ai that ranks agent skills for the next step using live session context.
Fallback
Show or change hmd's OmniRoute fallback-routing gate — whether Heimdall may route work to a third-party model,
Cheap LLM Router
Save Claude/Codex subscription tokens by routing context-only file reads through a cheaper LLM (DeepSeek, Kimi
CodeJoust
Pit AI coding agents against the same bug. Score them on tests, diff, cost, and time — pick the winning patch.
Multi Model Router
Route AI tasks to the best model using a single OpenAI-compatible API key. This Claude Code skill selects, con
Related Agents
Miner Medium
Delegate worker pinned to MEDIUM reasoning effort. Calibrated for C-coding since the 2026-07-07 refit (the C6
Shared Agent Protocols
Injected into agent context at spawn time. Do NOT add @decision annotations here — they waste agent tokens.
Loop Orchestrator
Route reflex-loop execution to the next role based on run state, stop guards, and last evaluation. Use proacti