home / mcp / langflow doc q&a mcp server
A Model Context Protocol server for document Q&A powered by Langflow . It demonstrates core MCP concepts by providing a simple interface to query documents through a Langflow backend.
Configuration
View docs{
"mcpServers": {
"gongrzhe-langflow-doc-qa-server": {
"command": "node",
"args": [
"/path/to/doc-qa-server/build/index.js"
],
"env": {
"API_ENDPOINT": "http://127.0.0.1:7860/api/v1/run/480ec7b3-29d2-4caa-b03b-e74118f35fac"
}
}
}
}You have a TypeScript MCP server that enables document Q&A through a Langflow backend. It exposes a simple query tool, letting you ask questions about documents and receive answers powered by your Langflow flow. This server is designed to be easy to run locally and integrated with MCP clients like Claude Desktop.
You use this server by connecting an MCP client to its local runtime. Start the server, then configure your client to point to the provided API endpoint. You can query documents via the query_docs tool, which sends your question to Langflow and returns the generated answer from the backend.
Prerequisites you need before installing this server are Node.js and npm. You will also need access to a Langflow instance that exposes a Run API endpoint.
# 1) Install dependencies
npm install
# 2) Build the TypeScript server
npm run build
# 3) (Optional) Enable auto-rebuild during development
npm run watchThe server is designed to run as an stdio MCP server. It uses a local Node process and requires an API endpoint to call Langflow. You will set the API endpoint in the server’s environment configuration so the MCP client can reach the backend.
When configuring Claude Desktop or another MCP client, supply the stdio configuration shown in the code block below. This starts the server and passes the required API endpoint for Langflow.
Query the document Q&A system by sending a user query to the Langflow backend and receiving a structured answer.