> For clean Markdown content of this page, append .md to this URL. For the complete documentation index, see https://learning.postman.com/llms.txt. For full content including API reference and SDK examples, see https://learning.postman.com/llms-full.txt.

# Use Postman to interact with your generated MCP server and an AI model

You can use Postman to interact with your generated MCP server locally and an AI model of your choice.

## Interact with your MCP server and an AI model

To interact with your MCP server and an AI model, do the following:

1. If you haven't already, [set up and start your MCP server](/docs/postman-api-network/showcase/publish/mcp-servers/set-up-start/).

2. [Create a new MCP request](/docs/use/send-requests/protocols/mcp-requests/create/#create-a-new-mcp-request) and [save it to a collection](/docs/use/send-requests/protocols/mcp-requests/create/#save-your-new-mcp-request). When you're prompted, choose your server's communication method:

   * To interact with your [standard input and output](https://modelcontextprotocol.io/docs/concepts/transports#standard-input%2Foutput-stdio) server, choose **STDIO** and enter your server's command. For example, enter:

     ```bash
     node /path/to/your-mcp-server/mcpServer.js
     ```

   * To interact with your [streamable HTTP](https://modelcontextprotocol.io/docs/concepts/transports#streamable-http) server, choose **HTTP** and enter your server's local URL. For example, enter:

     ```bash
     http://localhost:3001/mcp
     ```

3. [Use your MCP request to interact with your MCP server](/docs/use/send-requests/protocols/mcp-requests/interact/). You can use your request to experiment, test, and evaluate your MCP server.

4. [Create a new AI request](/docs/use/send-requests/protocols/ai-requests/create/#create-a-new-ai-request) and [save it to a collection](/docs/use/send-requests/protocols/ai-requests/create/#save-your-new-ai-request). Save your new AI request to the same workspace you saved your MCP request.

5. [Add your MCP request to your AI request](/docs/use/send-requests/protocols/ai-requests/add-mcp-servers/#add-the-mcp-server). When you're prompted, enter your server's command (STDIO) or URL (streamable HTTP), and search for the MCP request you created in the previous steps.

6. [Use your AI request to interact with your MCP server](/docs/use/send-requests/protocols/ai-requests/interact/). You can use your request to experiment, test, and evaluate AI models.

To learn more, see [Create MCP requests and add them to your collections](/docs/use/send-requests/protocols/mcp-requests/overview/) and [Create AI requests and add them to your collections](/docs/use/send-requests/protocols/ai-requests/overview/).