Skip to main content
Connect a CrewAI agent to Atlan over Streamable HTTP and make registry tools available to its assigned tasks.
Preview. This path uses CrewAI’s structured MCPServerHTTP configuration.

​
Prerequisites

  • A Python project using a current CrewAI release with MCP support.
  • CrewAI’s MCP dependencies resolved and recorded in the project lockfile.
  • The Atlan MCP endpoint: https://agentgateway.atlan.engineering/mcp.
  • OAuth support in the runtime.
  • Workspace access to the tools the crew should use.

​
Setup

  1. Configure an MCPServerHTTP connection for the Atlan MCP endpoint.
  2. Configure the runtime’s OAuth sign-in flow for the connection.
  3. Add the connection to the CrewAI agent that needs Atlan tools.
  4. Begin with a read-only task and a narrow tool set.
Keep credentials out of source code and endpoint URLs.

​
Expected result

CrewAI discovers the tools returned by Atlan and exposes them to the agent for its assigned tasks.

​
Verification

Run a small read-only task with verbose logging enabled. Confirm that CrewAI connects to the server, discovers the expected tool names, and completes without an authentication error.

​
Limitations

  • This integration is in Preview and is not enabled for every Atlan account.
  • Cached tool lists can outlive a permission change until the connection is rebuilt.
  • CrewAI may continue a task when an MCP server is unavailable, depending on the surrounding crew configuration.
  • Tool calls follow the authenticated Atlan identity and workspace.
See the CrewAI MCP guide for transport options, authentication, tool filtering, caching, and error behavior.