Skip to Content

MCP

MCP is an open protocol that standardizes how applications provide context to LLMs. You can think of MCP as the USB-C port for AI applications. Just like USB-C provides a standardized way to connect devices to various peripherals and accessories, MCP offers a standardized way to connect AI models with different data sources and tools.
— from anthropic

⚠️

For better tool invocation results, please use a base model that supports instruction following.

Starting from v0.1.0, the MCP client has been integrated natively. You can easily configure MCP servers via the MCP Management section in Settings.

Currently, only Stream MCP and Std MCP types are supported.

Once the MCP server is successfully configured, Revornix AI can interact with you using your entire tool library.

Developer

If you are interested in developing your own mcp server end, it is recommended to put the code in the api/mcp_router project group. The router folder is the interface group. For specific examples, please refer to the common and document examples in the router folder.

Last updated on