Jev directory
CommunityJev Agents & MCP
Jev MCP (Henrykey)
A self-contained Python FastMCP service exposing jev_choice, jev_score, jev_noul, and a batching jev_decide over Streamable HTTP, deployed behind a Caddy sidecar with a single shared bearer token.
- Reading time
- 2 min
- Published by
- Community
- Author
- henrykey
- Last verified
- Oct 2026
Used inJev for Software & DevTools
- Languages
- Python
- Capabilities
- Self-hostedRequires API key
Highlights
- Exposes jev_choice, jev_score, jev_noul, and jev_decide, which batches several judgments against one state in one upstream call.
- Python and FastMCP over Streamable HTTP; no database, no queue, and no OAuth server.
- Machine-to-machine auth is one shared bearer token compared in constant time; /health is unauthenticated and secret-free.
- Built for Tailscale/WireGuard-only access: Caddy terminates HTTPS with a locally trusted CA via tls internal.
- Successes return ok, request_id, and typed answers; Noul answers carry no confidence field because Jev returns none.
Watch out
An experimental, low-traffic internal service by the author's own description, not a product: access is over Tailscale/WireGuard with a self-signed certificate, and callers must be able to reach that address. The repository layout lists a LICENSE file but the page names no licence, so check the source before reuse.
Questions about this entry
Short answers drawn from the linked project or page.
Three bounded primitives — jev_choice, jev_score, and jev_noul — plus jev_decide, which batches several of them against one shared state in a single upstream TypeSafe call.
Copy .env.example to .env, set TYPESAFE_API_KEY and an MCP_AUTH_TOKEN generated with openssl rand -hex 32, then run docker compose up -d --build. Both the jev-mcp and jev-mcp-caddy containers should come up healthy, with /health reporting ok: true locally.
All requests to /mcp require Authorization: Bearer with the shared MCP_AUTH_TOKEN. GET /health is unauthenticated for container health checks and only returns ok plus safe fields, never secrets.
Because Jev does not return one for Noul, and the service does not invent values. If Jev fails or omits a field, the response reflects that instead of fabricating a value.
All tools return ok: true with request_id and results on success, or ok: false with request_id and an error object on failure — including upstream Jev failures. Nothing is silently fabricated.
The default deploy is not: it is Tailscale/WireGuard-only with a self-signed certificate, which is why curl needs -k. A separate public compose variant exists for a real domain with Let's Encrypt if needed.
It batches multiple typed questions against one state in a single upstream call, instead of forcing one round trip per judgment. The same evidence is shared across the typed questions in the batch.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: Jev in front of the HeyGen MCP
X: Playwright CLI + Jev vs Playwright MCP
X: A local MCP server that cuts computer use from 23s to 0.8s. Cartwright built Cortex, a local MCP server where…
X: A side panel that drives any site's WebMCP tools
More Jev resources like this
A Java/Spring Boot MCP server that exposes Jev as typed-judgment tools — jev_classify, jev_score, jev_check, jev_gate and jev_health — over HTTP (Streamable/SSE), for MCP clients that reach a running service instead of spawning a stdio subprocess.
Jev Agents & MCP#community#catalog-import#logicrw
Community
ashfaqbs/jev-mcp-springJava
ashfaqbs/jev-mcp-springJavaA plugin-based agent harness (FastAPI + Vue 3) bundling a chat GUI, multi-provider model access, an MCP client and marketplace, skills, workflow studio, sandboxed execution and a Tauri desktop shell. It integrates TypeSafe AI's Jev for structured Choice, Score and Noul decisions.
Jev Agents & MCP#community#github#agents
Community
andyqiuqiubo/andy-harnessTypeScript
andyqiuqiubo/andy-harnessTypeScriptA dependency-free MCP server that exposes TypeSafe's Jev to Claude Code, Codex, Cursor, Hermes, and other agents through the ask_jev and list_jev_models tools, so your code gets typed decisions with probabilities instead of prose to parse.
Jev Agents & MCP#community#catalog-import#logicrw
Community
codeia-academy/jev-mcpPython
codeia-academy/jev-mcpPythonA Python decision gateway and stdio MCP server that exposes Jev through OpenRouter's alpha decisions endpoint. Five tools cover yes/no checks, classification, scoring, multi-question evaluation and health, with bounded retries, validated answers and local diagnostics.
Jev Agents & MCP#community#mcp#openrouter
Community
ctmx/openrouter-jev-mcpPython
ctmx/openrouter-jev-mcpPythonUse Jev from Claude Code, Codex CLI, Claude Desktop or any MCP client, paid for by free Netlify credits and locked with a key only you have: a key-locked proxy forwards through Netlify AI Gateway and an MCP server exposes jev_evaluate and jev_models.
Jev Agents & MCP#community#catalog-import#logicrw
Community
demircigoksu/jev-netlify-mcpPython
demircigoksu/jev-netlify-mcpPythonA lightweight self-hosted System 1 decision daemon and MCP server for AI agents and multi-agent swarms: it evaluates state text against typed choice, score and noul questions in one forward pass, with Laya, Ollama and OpenAI-compatible backends.
Jev Agents & MCP#community#github#mcp
Community
axiomantic/local-systemonePython
axiomantic/local-systemonePythonFrom 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.