Skip to main content
The MCP (Model Context Protocol) integration lets you connect tools to and from Relevance AI. Give your agents access to tools hosted on external MCP servers, or expose your Relevance AI tools to services like Claude and Cursor. This means you can centralize your tool management in one place while extending their reach across your workflow — or pull in capabilities from other platforms without rebuilding them. There are two ways to use MCP with Relevance AI:
  1. External MCP server → Relevance AI Agent: Connect a remote MCP server to your agent, giving it access to tools from another service.
  2. Relevance AI tools → External service: Expose your Relevance AI tools to an external MCP-compatible client like Claude or Cursor.

​
Setting Up MCP Server Integration

​
For Agents

Agents can connect to and run tools from remote MCP servers.
  1. Open the Agent you want to connect
  2. Navigate to the “Tools” tab in the agent builder
  3. Click “Add MCP”
  4. Click “Connect my own”
  5. Add the URL of the remote hosted MCP server
  6. Label your MCP Server connection
  7. Add any authentication details required for the MCP server
Once connected, the agent will automatically fetch and display the available tools from the remote MCP server. These tools can be referenced in the agent’s core instructions and will appear in the conversation task view. When a tool is run, the agent communicates with the remote MCP server and returns the result directly in the conversation.
Local MCP server support (e.g. JSON configuration) is not available.

​
For Tools

You can access and use your Relevance AI tools via an MCP Server connection.
  1. Go to your Relevance AI workspace
  2. Click “More” on the left-side navigation menu, navigate to “MCP servers”
  3. Select the tools you want to share with external clients
  4. Click “Generate MCP Server Configuration”
  5. Copy your MCP configuration into your preferred client
For Cursor: Open Cursor settings → MCP tab → “Add new MCP server” → Paste into your mcp.json file

​
Client-Specific Setup

​
Cursor

To use your Relevance AI tools in Cursor:
  1. Open Cursor settings
  2. Navigate to the MCP tab
  3. Click “Add new MCP server”
  4. Paste your generated MCP configuration into your mcp.json file
  5. Save the configuration

​
Claude

To use your Relevance AI tools in Claude:
  1. Access Claude’s settings
  2. Look for the tool or API integration section
  3. Add a new MCP server
  4. Paste your generated MCP configuration
  5. Save the settings

​
Best Practices

  • Select tools strategically: Only expose tools that would be useful in external contexts
  • Test the integration: After setting up, test each tool to ensure it works as expected
  • Update regularly: When you update tools in Relevance AI, regenerate your MCP configuration
  • Monitor usage: Keep track of how your tools are being used across different platforms

​
Security Considerations

The MCP Server integration maintains the security of your tools while making them accessible to external clients. Authentication is handled through secure tokens, ensuring that only authorized users can access your tools. All communication between Relevance AI and external clients is encrypted to protect your data.

​
Troubleshooting

If you encounter issues with the MCP Server integration:
  • Verify that your MCP configuration is correctly copied into the external client
  • Ensure the external client supports the MCP standard
  • Check that the selected tools are compatible with the external client
  • Confirm your authentication credentials are valid
  • Regenerate the MCP configuration if necessary
Tool Templates - Create powerful tools that can be used both within Relevance AI and through the MCP Server integration. Create a Tool - Learn how to build custom tools that can be shared across your AI ecosystem. Customize Tools - Optimize your tools for better performance across different AI clients.

​
Conclusion

The MCP Server integration is a significant step forward in creating a unified AI ecosystem. By allowing your Relevance AI tools to be used in external AI clients, you can maintain consistency across platforms while leveraging the unique strengths of each environment. This integration maximizes your investment in tool development and ensures that your specialized capabilities are available wherever you work with AI.