mirror of
https://github.com/langbot-app/LangBot.git
synced 2026-10-29 20:05:25 +08:00
Xiaoyu SuandJunyan Chin
310c566148
* 重构文档结构：移除 docs 目录，调整文件组织结构 * Reorder "插件 SDK API" section in docs.json * Refactor and clarify various documentation sections * Refine warnings and info boxes across multiple docs * Refactor and simplify Japanese documentation content * Update package.json to include mintlify as a dependency and add pnpm-lock.yaml for dependency management * Update blog link and add description in docs.json --------- Co-authored-by: Junyan Chin <[email protected]>
--- title: "Using MCP Servers" --- LangBot supports using tools from MCP servers. You can add MCP servers to leverage the tool calling capabilities in large language models for better handling of LLM requests. <Info title="- MCP service supports HTTP Streamable, SSE and stdio modes."> - MCP servers recommended acquisition channel: [Composio MCP](https://mcp.composio.dev/) </Info> ## Enable Function Calling for Models As shown in the image, enable the `Function Calling` feature for the model. <img width="400px" src="/images/zh/deploy/mcp/mcp_01.png" alt="enable function call" /> ## Create MCP Server Click `Plugin Management`, then `MCP Management`, as shown: <img width="1000px" src="/images/zh/deploy/mcp/mcp02.png" alt="server page" /> Click `Add` in the top right corner, then `Create MCP Server`, <img width="400px" src="/images/zh/deploy/mcp/mcp03.png" alt="create new mcp server" /> For example, to use the Fetch HTML server, fill in the corresponding URL, click `Test` to test the MCP server, and click `Submit` to add the server. After adding, click the switch on the card to connect to the MCP server. Click the card to view the MCP server details: <img width="400px" src="/images/zh/deploy/mcp/mcp05.png" alt="detailed server" /> ## Effect You need to select `Local Agent` as the runner in the pipeline settings and use a model that supports function calling. <img width="1000px" src="/images/zh/deploy/mcp/mcp04.png" alt="dialog with mcp" />