Wikipedia Summarizer MCP Server
@codingaslu
An MCP (Model Context Protocol) server that fetches and summarizes Wikipedia articles using Ollama LLMs, accessible via both command-line and Streamlit interfaces. Perfect for quickly extracting key information from Wikipedia without reading entire articles.
Overview
What is Wikipedia Summarizer MCP Server?
It is an MCP server that fetches Wikipedia articles and summarizes them using local Ollama LLM models. It includes a command-line client and a Streamlit web interface for easy interaction.
How to use Wikipedia Summarizer MCP Server?
Install Python 3.8+ and Ollama (with the deepseek-r1:1.5b model), clone the repository, and install dependencies with uv pip install -r requirements.txt. Start the server with uv run -- ollama_server.py (available at http://localhost:8000/sse). Use the command-line client with uv run -- updated_client.py <server_url> <wikipedia_url> or launch the Streamlit interface with uv run -- streamlit run streamlit_new.py, then enter the server URL and article URL in the browser.
Key features of Wikipedia Summarizer MCP Server
- MCP server providing a
summarize_wikipedia_articletool - Command-line client for direct summarization requests
- Streamlit web interface for interactive use
- Uses Ollama LLM models (default
deepseek-r1:1.5b) - Fetches Wikipedia article content and converts to markdown
Use cases of Wikipedia Summarizer MCP Server
- Summarize any Wikipedia article using a local LLM
- Automate article summarization via command-line client
- Provide a web UI for non-technical users to get summaries
FAQ from Wikipedia Summarizer MCP Server
What are the prerequisites for running the server?
Python 3.8+, Ollama installed and running locally with the deepseek-r1:1.5b model, and an internet connection to fetch Wikipedia articles.
How do I install and run the server?
Clone the repository, run uv pip install -r requirements.txt, then start the server with uv run -- ollama_server.py. The server listens at http://localhost:8000/sse.
How do I use the command-line client?
Run uv run -- updated_client.py http://localhost:8000/sse https://en.wikipedia.org/wiki/Python_(programming_language) (replace the URLs as needed).
How do I use the Streamlit interface?
Run uv run -- streamlit run streamlit_new.py, open the provided URL in a browser, enter the MCP server URL and a Wikipedia article URL, then click "Fetch and Summarize Article".
What Ollama model does the server use by default?
It uses the deepseek-r1:1.5b model by default, but you can change it in ollama_server.py.
