Model Context Protocol (MCP)

This release contains access to DataStax Langflow as a Tech Preview. The Tech Preview is intended for you to view the capabilities of DataStax Langflow, and to provide your feedback to the product team. You are permitted to use the information only for evaluation purposes and not for use in a production environment. IBM provides the information without obligation of support and "as is" without warranty of any kind.

DataStax Langflow integrates with the Model Context Protocol (MCP) as both an MCP server and an MCP client.

This page describes the authentication and configuration requirements specific to DataStax Langflow. For more information about MCP in Langflow, see Use Langflow as an MCP server and Use Langflow as an MCP client.

Prerequisites

Sign in or create an Astra account
Create an Astra application token with the Langflow Administrator or Organization Administrator role
Your Astra organization ID
A flow that you want to expose with MCP

Authentication

When connecting to DataStax Langflow with MCP, you must provide two authentication credentials:

Application token: Your Astra application token
Organization ID: The ID of the organization associated with your application token and the flow that you want to run.

You must include both credentials in the request headers.

DataStax Langflow only supports header-based authentication with Astra application tokens and organization IDs. If you want to use another authentication method, you must use Langflow OSS. For more information, see Use Langflow as an MCP server.

Connect MCP clients to DataStax Langflow

You can connect any MCP-compatible client application such as Cursor, Claude Desktop, and Windsurf to DataStax Langflow using the MCP remote server configuration:

Install any MCP-compatible client.

The process is generally the same for all clients, with slight differences in client-specific details like file names and configuration locations.
In your client, add a new MCP server using the client’s UI or configuration file.

For example, in Cursor, go to Cursor Settings, select MCP, and then click Add New Global MCP Server to open Cursor’s global mcp.json configuration file.
Get your MCP endpoint URL and configuration.

In Langflow, on the Projects page, click the MCP Server tab.

Copy the code snippet for your operating system. The MCP Server tab automatically populates the placeholder values with your project’s actual values.

{
  "mcpServers": {
    "PROJECT_NAME": {
      "command": "uvx",
      "args": [
        "mcp-proxy",
        "https://DATASTAX_LANGFLOW_URL/lf/LANGFLOW_TENANT_ID/api/v1/mcp/project/PROJECT_UUID/streamable",
        "--transport",
        "streamablehttp",
        "--headers",
        "X-DataStax-Current-Org",
        "DATASTAX_ORG_ID",
        "--headers",
        "Authorization",
        "Bearer AstraCS:APPLICATION_TOKEN"
      ],
      "enabled": true
    }
  }
}

{
  "mcpServers": {
    "PROJECT_NAME": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote@latest",
        "https://DATASTAX_LANGFLOW_URL/lf/LANGFLOW_TENANT_ID/api/v1/mcp/project/PROJECT_UUID/streamable",
        "--header",
        "X-DataStax-Current-Org: DATASTAX_ORG_ID",
        "--header",
        "Authorization: Bearer AstraCS:APPLICATION_TOKEN"
      ]
    }
  }
}

Save and close your client’s MCP configuration file.
Confirm that your DataStax Langflow MCP server is on the client’s list of MCP servers. If necessary, restart your client to apply the modified configuration file.

Connect MCP servers to DataStax Langflow

You can use DataStax Langflow as an MCP client to connect to external MCP servers and use their tools within your Langflow flows.

The MCP Tools component functionality in DataStax Langflow is the same as in Langflow OSS. No additional authentication or configuration is required beyond what’s needed for the external MCP server you’re connecting to.

For more information, see Use Langflow as an MCP client.

Model Context Protocol (MCP)

Prerequisites

Authentication

Connect MCP clients to DataStax Langflow

Connect MCP servers to DataStax Langflow

Was this helpful?

Give Feedback