概要
What is Wikipedia MCP Server?
A server for searching and analyzing Wikipedia articles using AI models, accessed via HTTP endpoints and Server-Sent Events.
How to use Wikipedia MCP Server?
Install dependencies with pip install -r requirements.txt, set the OPENROUTER_API_KEY environment variable, then run python wiki_mcp_server.py. The server starts on http://localhost:8000. Use the /search, /evaluate, /analyze, and /stats endpoints with POST or GET requests.
Key features of Wikipedia MCP Server
- Search Wikipedia articles with AI‑powered relevance scoring.
- Evaluate a single article’s relevance using an AI model.
- Analyze article content with AI‑generated insights.
- Server‑Sent Events (SSE) stream for real‑time search results.
- Supports multiple AI models: GPT‑3.5‑Turbo, GPT‑4, Claude‑2, Mistral‑7B.
- Built‑in server statistics and error tracking.
Use cases of Wikipedia MCP Server
- Automatically retrieve and rank Wikipedia articles on a given topic.
- Integrate Wikipedia knowledge into AI chatbots or research assistants.
- Evaluate the relevance of a specific article snippet for a project.
- Combine article analysis with other MCP tools in an AI pipeline.
FAQ from Wikipedia MCP Server
What API key is required?
The server requires an OpenRouter API key set via the OPENROUTER_API_KEY environment variable.
What AI models are available?
The server supports gpt-3.5-turbo (default), gpt-4, claude-2, and mistral-7b.
How do I run the server?
Install dependencies, export the API key, and run python wiki_mcp_server.py. The server will listen on port 8000.
What endpoints does the server provide?
There are four endpoints: /search (POST, SSE), /evaluate (POST, JSON), /analyze (POST, JSON), and /stats (GET, JSON).
Can I configure server timeouts or connections?
Yes, optional environment variables SERVER_TIMEOUT, KEEPALIVE_TIMEOUT, and MAX_CONNECTIONS can be set.
