Extended the anthropic mcp tutorial to support openai and customGPT connection through http-bridge
Overview
What is customGPT-to-mcp-server-example?
This project is a repository that extends the Model Context Protocol (MCP) tutorial to support connections between OpenAI's custom GPTs and MCP servers through an HTTP bridge.
How to use customGPT-to-mcp-server-example?
To use this project, follow these steps: 1) Start the MCP Weather Server, 2) Start the HTTP Bridge, 3) Expose your local server to the internet using ngrok, 4) Prepare your OpenAPI schema, 5) Configure your custom GPT in the GPT Builder, and 6) Test your custom GPT.
Key features of customGPT-to-mcp-server-example?
- Integration methods for custom GPTs via HTTP API bridge.
- Support for both local and remote MCP servers.
- Detailed setup instructions for connecting custom GPTs to MCP servers.
Use cases of customGPT-to-mcp-server-example?
- Connecting custom GPTs to weather data services.
- Building LLM-powered chatbots that interact with MCP servers.
- Creating REST API servers for custom applications.
FAQ from customGPT-to-mcp-server-example?
- Can I use this project with any MCP server?
Yes! The HTTP bridge can connect to any MCP server that is accessible from your machine.
- Is there a public URL requirement?
Yes, OpenAI requires a public HTTPS endpoint for Actions, which can be achieved using ngrok.
- What if I encounter issues with schema import?
If importing from URL fails, you can copy the JSON and paste it manually.