Enables MCP-speaking hosts to ask typed judgment questions — yes/no, choice, and score — and receive calibrated probabilities instead of prose. Runs offline by default against a synthetic provider, with an opt-in live route whose egress is disclosed before any data leaves.
Enables MCP clients to get machine-readable yes/no, choice, and score decisions from TypeSafe Jev, bridging fast classification to Cursor and other clients.
Exposes bounded TypeSafe Jev/System One decision primitives as MCP tools, enabling agents to make choices, scores, yes/no judgments, and batched decisions over remote HTTPS with shared state.
Enables MCP-capable agents to run TypeSafe's Jev judgment model as typed yes/no, choice, and score tools, with calibrated probabilities, confidence thresholds, escalation for uncertain or non-judgment tasks, and an optional action gate that fails open.
Exposes three stdio tools that answer typed questions — yes/no, choice, and score — with calibrated probabilities instead of prose, letting any MCP host query a judgment model without a chat interface. Runs fully offline on a synthetic mock by default, with named, per-feature egress disclosure and redaction only when a live provider is explicitly configured.