Skip to main content

MCP Integration

Definity provides Model Context Protocol (MCP) integration, allowing AI agents to interact with your data observability platform programmatically.

What is MCP?​

The Model Context Protocol (MCP) is an open standard that enables AI assistants to securely connect to various data sources and tools. With Definity's MCP server, AI agents can:

  • Query pipeline metadata and execution history
  • Retrieve data quality metrics and anomalies
  • Access lineage information
  • Monitor task statuses and performance

Prerequisites​

  • Integration API token (generate in Definity UI under User → Generate API Token)
  • MCP-compatible AI agent (e.g., Claude Desktop, Cline, etc.)
  • Node.js and npm installed on your system

Configuration​

Claude Desktop​

Add the Definity MCP server to your Claude Desktop configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

Windows: %APPDATA%\Claude\claude_desktop_config.json

{
"mcpServers": {
"definity": {
"command": "npx",
"args": [
"mcp-remote",
"https://app.definity.run/api/mcp",
"--header",
"Authorization:${AUTH_HEADER}"
],
"env": {
"AUTH_HEADER": "Bearer your_integration_token_here"
}
}
}
}

Other MCP Clients​

For other MCP-compatible clients, connect to the Definity MCP endpoint:

# Server endpoint
https://app.definity.run/api/mcp/

# Authentication header
Authorization: Bearer your_integration_token_here

Testing with MCP Inspector​

You can test your Definity MCP connection and explore available tools using the MCP Inspector:

Installation​

npm install -g @modelcontextprotocol/inspector

Usage​

Run the inspector with your Definity configuration:

npx @modelcontextprotocol/inspector mcp-remote https://app.definity.run/api/mcp --header "Authorization:Bearer your_integration_token_here"

The inspector will open a web interface where you can:

  • View all available tools and their schemas
  • Test tool calls with different parameters
  • See request/response payloads
  • Debug connection issues

This is useful for:

  • Verifying your API token has the correct permissions
  • Understanding tool parameters before using them in your agent
  • Troubleshooting connection or authentication problems