Jev MCP server
The Jev MCP server drops a typed decision tool into any Model Context Protocol client. Instead of the client guessing and parsing, it calls one tool and gets back a locked answer — one of your options, calibrated probabilities, and confidence.
MCP lets an agent call external tools; Jev is the tool you want for any yes/no, pick-one, or classify decision. Jev is TypeSafe AI's System One model, so the tool returns a typed structured value — never free text the client has to re-interpret — in roughly 70–500ms, in a single non-autoregressive pass.
Run it
// Run the Jev MCP server straight from GitHub — no install step:
npx github:codaaiteam/jev-mcp
// Set your key in the environment first:
export JEV_API_KEY=jv_live_...Point your MCP client (Claude Code, Claude Desktop, or any MCP host) at that command. The server registers a decide tool the model can call whenever it needs a typed, calibrated answer instead of a guess.
Why an MCP decision tool
- The client gets one option locked to your list — no invalid labels, 0% structured-output error
- Calibrated confidence (RLCD) the agent can gate on
- One round trip, no parsing on the client side
- Runs over npx from github: — nothing to build or host
Under the hood
The MCP server is a thin wrapper over the same endpoint everything else uses: POST https://jevtypesafeai.com/api/v1/decide with a jv_live_ key (official upstream POST https://api.typesafe.ai/v1/systemone). See the full integration guide for config options and tool schemas.
FAQ
What does the Jev MCP server give my agent?
A single decide tool. The model calls it with state and typed questions and gets back one option per question with calibrated confidence — a reliable decision primitive instead of a free-text guess it has to parse.
Do I need to install anything?
No. Run npx github:codaaiteam/jev-mcp with your JEV_API_KEY set, and point any MCP host at the command. There's nothing to build or self-host.
How is this different from the REST API?
It's the same /api/v1/decide endpoint under the hood — the MCP server just exposes it as a native tool so MCP clients can call it without you writing the HTTP glue.
See also: MCP integration guide · Jev for AI agents · Docs
Add the Jev MCP tool
Grab a jv_live_ key and give your MCP client a typed decision tool in one command.