joaowinderfeldbussolotto MCP Websearch Server

MPC Docs Server

This is a simple MCP (Model Context Protocol) server for retrieving information from the official documentation of Langchain, Llama Index, and OpenAI. It provides a tool that can be used by MCP-compatible applications to search and retrieve relevant documentation snippets.

Features

• Documentation Retrieval: Fetches content from the official documentation of Langchain, Llama Index, and OpenAI.
• MCP Compatibility: Implements an MCP server, allowing it to be easily integrated with other MCP-compatible applications.
• Simple Tool: Exposes a get_docs tool that accepts a query and library name, returning relevant documentation snippets.

How It Works

graph LR
    Client[MCP Client] -->|Calls tools| Server[MCP Server]
    Server -->|Searches web for docs| Serper[Serper API]
    Serper -->|Returns search results| Server
    Server -->|Returns documentation| Client

Getting Started

Installing uv Package Manager

On MacOS/Linux:

curl -LsSf https://astral.sh/uv/install.sh | sh

Make sure to restart your terminal afterwards to ensure that the uv command gets picked up.

Project Setup

Create and initialize the project:

# Create a new directory for our project
uv init mcp-server
cd mcp-server

# Create virtual environment and activate it
uv venv
source .venv`/bin/activate`  # On Windows use: .venv\Scripts\activate

# Install dependencies
uv add "mcp[cli]" httpx python-dotenv bs4

Environment Variables

Create a .env file in the root directory and add the following:

SERPER_API_KEY=YOUR_SERPER_API_KEY

You'll need a SERPER API key to use the web search functionality. You can obtain one from Serper.dev. We are using the Serper API to search the web for relevant documentation.

Running the Server

Start the MCP server:

uv run main.py

The server will start and be ready to accept connections.