esp

The esp-mcp-server exposes a Model Context Protocol (MCP) endpoint for Espressif AI-assisted development tools. You can run it locally via Docker and then point your IDEs or tools to the MCP endpoint at http://127.0.0.1:8080/mcp. The server is designed to help IDEs retrieve model contexts, documentation, and support queries related to ESP chips, ESP-IDF, and related Espressif materials. The supported integration example shows how to configure an IDE cursor and VSCode extension to connect to the MCP endpoint, typically supplying an API key in the environment configuration.

To use the server, start the container and connect to the /mcp path from your tools. For example, in a cursor-based integration you would reference the endpoint http://127.0.0.1:8080/mcp and provide any required API key in the environment variables section of your tool configuration. The server then answers queries, serves MCP streams, and facilitates context-aware assistance for Espressif developments.

Prerequisites:

• Docker installed on your machine
• Internet access to pull the image

Installation steps:

1. Pull and run the ESP MCP server container:

docker run -p 8080:8080 kolipakakondal/esp-mcp-server:latest

This starts the server and exposes the MCP endpoint at http://127.0.0.1:8080/mcp.

2. Verify the endpoint is reachable:

curl http://127.0.0.1:8080/mcp

You should receive a response indicating the MCP service is available.

3. Configure your IDE/tools to point to the MCP endpoint as shown in the README examples, including any required environment variables such as API_KEY if your workflow demands it.

Tips:

• Ensure Docker is running and that port 8080 is not blocked by a firewall.
• If you change the host port, update your tool configurations accordingly (e.g., http://localhost:PORT/mcp).
• The server supports integration with tools like VSCode and Cursor by specifying the /mcp endpoint and supplying an API_KEY in the env section if required by your workflow.
• If you encounter connection issues, check container logs with docker ps and docker logs <container_id> for errors.