OpenClaw Consensus

Overview

OpenClaw Consensus is a Model Context Protocol (MCP) server: 9-LLM consensus + disagreement scoring + cheapest-route picks to fight hallucinations. In practice that means any MCP-compatible AI assistant can call OpenClaw Consensus's tools directly — the model decides when to use them in a conversation or agent run.

It runs locally: the Python package openclaw-consensus-mcp (via uvx) speaks MCP over stdio on your machine. Configuration happens through environment variables — you'll need RAPIDAPI_KEY. The current release is v0.1.1, published under the MIT license, with source at MICONNM/openclaw-consensus-mcp on GitHub.

OpenClaw Consensus is listed under AI & Agents on mcp.site and works with any MCP client — Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed and the rest of the ecosystem — using the install snippets below. If you maintain OpenClaw Consensus for MICONNM, claim this listing to verify ownership, earn the Verified badge, and keep the details current.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "openclaw_consensus": {
      "command": "uvx",
      "args": [
        "openclaw-consensus-mcp"
      ],
      "env": {
        "RAPIDAPI_KEY": "<rapidapi-key>"
      }
    }
  }
}

Requires RAPIDAPI_KEY