Relevanceai MCP Server

This server allows you to use your Relevance AI tools (or studios) within MCP-compatible clients like Cursor or Claude.

Usage with NPX

The easiest way to run this server is using npx. You need Node.js and npm installed.

In your MCP client's configuration (e.g., Cursor's settings), set the "MCP Server Command" to the following. First, ensure you have the RELEVANCE_AUTH_TOKEN environment variable set with your Relevance AI API key. Then, use the command, replacing the placeholders with your actual values:

npx @relevanceai/relevanceai-mcp-server@latest --region YOUR_RELEVANCE_REGION --tools YOUR_TOOL_ID1,YOUR_TOOL_ID2

Environment Variable:

• RELEVANCE_AUTH_TOKEN: Your Relevance AI API key (create one under Account -> Integrations in your Relevance AI dashboard). This must be set in your environment.

Command-line Arguments:

• --region YOUR_RELEVANCE_REGION: Your Relevance AI region (e.g., 1e3042).
• --tools YOUR_TOOL_ID1,YOUR_TOOL_ID2: A comma-separated list of the Tool IDs you want to expose.

Example:

Assuming RELEVANCE_AUTH_TOKEN is set in your environment (e.g., export RELEVANCE_AUTH_TOKEN="abcdef12345"):

npx @relevanceai/relevanceai-mcp-server@latest --region 1e3042 --tools tooldId1,toolId2

Development Setup

If you want to contribute or modify the server:

1. Clone the repository: git clone https://github.com/RelevanceAI/relevanceai-mcp-server
2. Navigate to the directory: cd relevanceai-mcp-server
3. Install dependencies: npm install
4. Build the project: npm run build
5. Run locally (for testing): First, set the environment variable: export RELEVANCE_AUTH_TOKEN="YOUR_RELEVANCE_AUTH_TOKEN" Then, run the server with arguments: npm start -- --region <region> --tools <toolId1,toolId2>

Logging

The server uses console.error for logging. This is compatible with how MCP clients capture server output. Avoid using console.log as it can interfere with the MCP communication protocol.