Wikipedia Simple English

Access Wikipedia content, prioritizing Simple English with a fallback to regular English.

A Model Context Protocol (MCP) server that provides access to Wikipedia content, prioritizing Simple English Wikipedia with intelligent fallback to regular English Wikipedia when content is unavailable or of poor quality.

This MCP server acts as a bridge between AI assistants and Wikipedia, specifically designed to:

- Prioritize Simple English: Uses Simple English Wikipedia first for clearer, more accessible content
- Intelligent Fallback: Automatically falls back to regular English Wikipedia when needed
- Quality Detection: Evaluates content quality and switches to English Wikipedia for stub articles
- Human-Friendly: Designed with readability and usability in mind

- Search across Wikipedia articles
- Returns structured metadata and ranked list of relevant page titles
- Tries Simple English first, falls back to English
- Configurable result limits

- Retrieve Wikipedia page summaries/excerpts
- Lightweight, fast responses for quick overviews
- Structured metadata with content length and language indicators
- Automatic disambiguation handling

- Retrieve full Wikipedia page content
- Complete article text for detailed research
- Structured output with separate metadata and content blocks
- Quality-based language selection

- Python 3.12 or higher
- uvpackage manager

uvx git+https://github.com/bhubbb/mcp-se-wikipedia

This MCP server is designed to work with AI assistants that support the Model Context Protocol. Configure your AI assistant to connect to this server via stdio.

Example configuration for Claude Desktop:

{ "mcpServers": { "wikipedia-se": { "command": "uvx", "args": ["git+https://github.com/bhubbb/mcp-se-wikipedia"] } } }

{ "mcpServers": { "wikipedia-se": { "command": "uv", "args": ["run", "python", "/path/to/mcp-se-wikipedia/main.py"] } } }

To use full English Wikipedia exclusively (bypassing Simple English):

{ "mcpServers": { "wikipedia-se": { "command": "uvx", "args": ["git+https://github.com/bhubbb/mcp-se-wikipedia"], "env": { "WIKIPEDIA_MODE": "full" } } } }

The server behavior can be configured using environment variables:

Controls which Wikipedia version to use:

- simple(default): Prioritizes Simple English Wikipedia with automatic fallback to regular English
- full: Uses only regular English Wikipedia, bypassing Simple English entirely

# Default mode (Simple English first) uv run python main.py # Full Wikipedia mode only WIKIPEDIA_MODE=full uv run python main.py

- You need more detailed technical content
- You want consistent results from regular Wikipedia
- You're working with topics that have better coverage in full English Wikipedia

Search Wikipedia for articles with structured results.

- query(required): Search terms
- limit(optional): Max results to return (1-20, default: 10)

- Search metadata (Wikipedia version, query, results count, language code)
- List of matching article titles

{ "name": "search", "arguments": { "query": "solar system", "limit": 5 } }

Get the summary/excerpt of a Wikipedia page.

- title(required): Page title to retrieve summary for
- auto_suggest(optional): Auto-suggest similar titles if exact match not found (default: true)

- Summary metadata (title, Wikipedia version, language code, URL, content length)
- Page summary text

{ "name": "summary", "arguments": { "title": "Earth", "auto_suggest": true } }

Get the full content of a Wikipedia page.

- title(required): Page title to retrieve full content for
- auto_suggest(optional): Auto-suggest similar titles if exact match not found (default: true)

- Content metadata (title, Wikipedia version, language code, URL, content length)
- Complete article content

{ "name": "content", "arguments": { "title": "Earth", "auto_suggest": true } }

- Simple English First(default mode): All requests start with Simple English Wikipedia
- Quality Check: For content/summary requests, checks if content is substantial (>500 characters)
- Automatic Fallback: Switches to English Wikipedia if:

- Page doesn't exist in Simple English
- Content is too brief (likely a stub)
- Search returns no results

All tools return structured responses with separate blocks for:

- Metadata: Wikipedia version, language codes, URLs, content statistics
- Content: Search results, summaries, or full article content
- Options: Disambiguation choices when multiple pages match

This structure makes responses both human-readable and machine-parseable.

- Disambiguation: When multiple pages match, returns list of options
- Page Not Found: Clear error messages with suggestions
- Network Issues: Graceful error handling with informative messages

- Get simplified explanations of complex topics
- Access educational content in clearer language
- Research assistance with automatic complexity adjustment

- Source material for articles and explanations
- Fact-checking with reliable Wikipedia sources
- Research support with readability optimization

- Easier-to-read content for various reading levels
- Simplified language for non-native speakers
- Clear, concise information delivery

Ask your AI assistant:"Search Wikipedia for information about photosynthesis"
- Search Simple English Wikipedia for "photosynthesis"
- Return structured metadata (version, language code, result count)
- Provide a list of relevant articles
- Fall back to English Wikipedia if needed

Ask your AI assistant:"Get a summary of the Moon from Wikipedia"
- Try to fetch "Moon" summary from Simple English Wikipedia
- Return structured metadata (title, version, URL, content length)
- Provide the page summary
- Fall back to English Wikipedia if Simple English unavailable

Ask your AI assistant:"Get the full Wikipedia article about the Moon"
- Try to fetch complete "Moon" article from Simple English Wikipedia
- Check if the content is substantial enough
- Return structured metadata and full article content
- Fall back to English Wikipedia if needed

Ask your AI assistant:"Get information about Mercury"

If multiple pages match (planet Mercury, element Mercury, etc.), the server will return:

- Structured disambiguation metadata
- A list of options to choose from

mcp-se-wikipedia/ ├── main.py # Main MCP server implementation ├── pyproject.toml # Project dependencies and metadata ├── AGENT.md # This documentation └── .venv/ # Virtual environment (created by uv)

- mcp: Model Context Protocol framework
- wikipedia: Python Wikipedia API wrapper
- asyncio: Async/await support for MCP

The server can be easily customized by modifyingmain.py:

- Quality Threshold: Change the 500-character threshold for content quality
- Search Limits: Modify default and maximum search result limits
- Language Settings: Add support for other Wikipedia languages
- Content Filtering: Add custom content processing or filtering

uvx git+https://github.com/bhubbb/mcp-se-wikipedia

cd mcp-se-wikipedia uv run python main.py

The server communicates via JSON-RPC over stdin/stdout, so you'll need an MCP client or AI assistant to interact with it properly.
- Import Errors: Make sure all dependencies are installed withuv sync
- Connection Issues: Check network connectivity for Wikipedia API access
- Rate Limiting: Wikipedia has rate limits; the server handles this gracefully

Enable debug logging by modifying the logging level inmain.py:

logging.basicConfig(level=logging.DEBUG)

This is a simple, single-file implementation designed for clarity and ease of modification. Feel free to:

- Add new Wikipedia language support
- Implement additional content filtering
- Add caching for better performance
- Extend with additional Wikipedia features

This project uses the same license as its dependencies. The Wikipedia content accessed through this server is subject to Wikipedia's licensing terms.

Search global news using natural language. Webz.io News Search API returns the most relevant articles and content, with filters for source, country, language, date, sentiment, and category.

Fetch, convert, and search AWS documentation pages, with recommendations for related content.

Provides AI assistants with intelligent access to ML textbook content for creating accurate, source-grounded documentation.

一条工務店で家を建てた施主「ろれさん」のブログ記事と、YouTube/X/Instagram/Web から集めた約3,000件の家づくり Tips を横断検索できる MCP サーバ。すべての結果に出典URLが付きます。

A flexible service for searching and analyzing academic papers on arXiv.

Caesar is a free, keyless web search API for AI agents. Its remote MCP server exposes web_search (ranked results with citable provenance) and web_fetch (full pages as clean markdown), and works anonymously with no API key.

CatchAll is a web search API built for comprehensive event retrieval — not ranked results, but all matching records.

Creates a personal, always-current knowledge base for AI by indexing documentation from websites, GitHub, npm, PyPI, and local files.

Provides real-time access to documentation, library popularity data, and career insights using the Serper API.

GitHits MCP gives AI agents access to open-source code search, package docs, real-world examples, dependency metadata, changelogs, and vulnerability context for better software development decisions.