Skip to main content
MCP server CX Codex

clef-mcp — Claude Code MCP server: what it does, how to install

Runs the Clef-Flash 9B model locally to score agent choices. Returns strict JSON probability distributions offline without burning cloud tokens.

Install into Claude Code · Platform CLI All methods and platforms ↓
claude mcp add clef-mcp -- npx -y clef-mcp
Open on GitHub

What clef-mcp does

The server exposes Cloudflare's Clef-Flash 9B decision model via a single clef_decide tool. Agents supply task context and typed questions without requesting lengthy paragraphs. The engine outputs probabilities across candidate answers in one forward pass. Latency stays low.

Work happens on-device inside llama.cpp without external network calls. You can score up to 64 questions in a single request across discrete choices, scales, or binary checks. The state payload is read purely as data. Skipping the initial model download of roughly 6 GB leads straight to a MODEL_NOT_INSTALLED error.

What it helps with

  • Select the next agent action when diagnosing test failures
  • Score confidence levels before executing risky code edits
  • Route incoming issue tickets into team categories by urgency
  • Classify incident severity directly on the local machine

Example prompts for the agent

  • Use clef_decide to score whether to inspect the files or ask the user.
  • Run the incident-triage prompt on the current deployment logs.
  • Evaluate next actions and calculate probability scores for each option.

How to install clef-mcp

Platform CLI

claude mcp add clef-mcp -- npx -y clef-mcp

Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more

Questions about clef-mcp

How do I install clef-mcp in Claude Code?

Run claude mcp add clef-mcp -- npx -y clef-mcp in your terminal. Ensure you run npx clef-mcp install beforehand to download runtime binaries and model weights.

What does the clef_decide tool return?

It outputs a structured JSON object containing probability distributions for each asked question. The model evaluates options in a single pass without prose.

Does clef-mcp require internet access during execution?

Inference runs completely offline on your hardware using llama.cpp. A network connection is needed only during initial setup to download the 6 GB model file.

Similar MCP servers

All MCP servers for Agent setup: memory, context, orchestration →

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor