1999AZZAR/wikipedia-mcp-server

Wikipedia MCP Server

A comprehensive Model Context Protocol (MCP) server providing access to Wikipedia content with intelligent caching, batch operations, and advanced search capabilities for local development.

Certification

This MCP server is certified by MCP Review - your trusted platform for discovering and evaluating Model Context Protocol servers.

Features

Core MCP Tools

• search - Enhanced search with snippet control and pagination
• getPage - Full page content with configurable sections, images, links, and categories
• getPageById - Page retrieval by ID with the same enhancement options
• getPageSummary - Fast page summaries via the Wikipedia REST API
• random - Random article discovery
• pageLanguages - Lists available languages for a given page

Advanced MCP Tools

• batchSearch - Search multiple queries simultaneously for efficiency
• batchGetPages - Retrieve multiple pages at once with concurrency control
• searchNearby - Find Wikipedia articles near specific coordinates
• getPagesInCategory - Browse pages within Wikipedia categories

Performance & Reliability

• Intelligent Caching - Memory-based caching for improved performance
• Request Optimization - Endpoint rotation and intelligent routing
• Error Handling - Robust error handling with retry logic
• TypeScript - Fully typed with strict error handling
• Comprehensive Testing - Jest test suite with health validations

Installation

1. Clone or download this repository
2. Install dependencies:

   npm install
   

3. Build the project:

   npm run build
   

Configuration

Add the Wikipedia MCP server to your MCP client configuration:

{
  "mcpServers": {
    "wikipedia": {
      "command": "node",
      "args": ["/path/to/wikipedia-mcp-server/index.js"],
      "env": {
        "CACHE_MAX": "500",
        "CACHE_TTL": "300000",
        "DEFAULT_LANGUAGE": "en",
        "ENABLE_DEDUPLICATION": "true"
      }
    }
  }
}

Note: Replace /path/to/wikipedia-mcp-server with your actual path to the wikipedia-mcp-server directory.

Environment Variables

• CACHE_MAX: Maximum number of items in memory cache (default: 100)
• CACHE_TTL: Cache TTL in milliseconds (default: 300000 = 5 minutes)
• DEFAULT_LANGUAGE: Default Wikipedia language (default: en)
• ENABLE_DEDUPLICATION: Enable request deduplication (default: true)
• USER_AGENT: Custom user agent for Wikipedia API requests

Running Locally

Start the server using npm:

npm start

This starts the MCP server that communicates via stdio, perfect for integration with MCP clients like Cursor.

Available Tools

Core MCP Tools

• search - Enhanced search with snippet control and pagination
• getPage - Full page content with configurable sections, images, links, and categories
• getPageById - Page retrieval by ID with the same enhancement options
• getPageSummary - Fast page summaries via the Wikipedia REST API
• random - Random article discovery
• pageLanguages - Lists available languages for a given page

Advanced MCP Tools

• batchSearch - Search multiple queries simultaneously for efficiency
• batchGetPages - Retrieve multiple pages at once with concurrency control
• searchNearby - Find Wikipedia articles near specific coordinates
• getPagesInCategory - Browse pages within Wikipedia categories

Available Resources

Wikipedia MCP Server provides 7 specialized dynamic resources that offer cached Wikipedia data and intelligent content access. These are URI-based resources that can be accessed directly by constructing the appropriate URI pattern.

Note: These resources are dynamic and do not appear in standard resources/list calls. Access them directly using resources/read with the URI patterns below:

wikipedia://article/{title}/{lang}

Returns cached full content of a Wikipedia article with metadata and revision information.

Resource Details:

• Purpose: Access complete article content with local caching
• Benefits: Faster article loading and reduced API calls
• Use Cases: Article reading, content analysis, research workflows

Response Format:

{
  "title": "Artificial Intelligence",
  "lang": "en",
  "content": {
    "pageid": 1164,
    "ns": 0,
    "title": "Artificial Intelligence",
    "revisions": [...],
    "links": [...],
    "categories": [...]
  },
  "cached": false,
  "timestamp": "2025-11-02T17:09:14.866Z"
}

wikipedia://search/cache/{query}

Provides cached search results for Wikipedia queries with snippets and metadata.

Resource Details:

• Purpose: Cache search operations for improved performance
• Benefits: Fast repeated searches and reduced API usage
• Use Cases: Research queries, content discovery, topic exploration

Response Format:

{
  "query": "machine learning",
  "results": [
    {
      "pageid": 233488,
      "title": "Machine learning",
      "snippet": "Machine learning is a method of data analysis that automates analytical model building..."
    }
  ],
  "totalResults": 1500,
  "cached": false,
  "timestamp": "2025-11-02T17:09:14.866Z"
}

wikipedia://metadata/{title}

Offers comprehensive metadata for Wikipedia articles including categories, links, and references.

Resource Details:

• Purpose: Access rich metadata without full article content
• Benefits: Lightweight metadata queries and relationship discovery
• Use Cases: Citation analysis, category exploration, link analysis

Response Format:

{
  "title": "Machine Learning",
  "pageid": 233488,
  "ns": 0,
  "revid": 123456789,
  "lastModified": "2025-11-02T15:30:00Z",
  "categories": [
    { "ns": 14, "title": "Category:Machine learning" }
  ],
  "links": [
    { "ns": 0, "title": "Artificial intelligence" }
  ],
  "references": ["https://doi.org/10.1038/nature14539"],
  "cached": false,
  "timestamp": "2025-11-02T17:09:14.866Z"
}

wikipedia://categories/{title}

Provides category hierarchy and related articles for Wikipedia categories.

Resource Details:

• Purpose: Explore category structures and related content
• Benefits: Topic discovery and content organization
• Use Cases: Topic research, content curation, knowledge organization

Response Format:

{
  "category": "Machine learning",
  "memberCount": 150,
  "members": [
    { "pageid": 123, "ns": 0, "title": "Neural network" }
  ],
  "subcategories": [
    { "pageid": 456, "ns": 14, "title": "Category:Deep learning" }
  ],
  "articles": [
    { "pageid": 789, "ns": 0, "title": "Supervised learning" }
  ],
  "cached": false,
  "timestamp": "2025-11-02T17:09:14.866Z"
}

wikipedia://languages/{title}

Shows available language variants for a Wikipedia article.

Resource Details:

• Purpose: Discover multilingual content availability
• Benefits: Language coverage analysis and translation discovery
• Use Cases: Multilingual research, translation workflows, content localization

Response Format:

{
  "title": "Machine Learning",
  "languageCount": 47,
  "languages": [
    {
      "lang": "es",
      "title": "Aprendizaje automático",
      "url": "https://es.wikipedia.org/wiki/Aprendizaje_automático"
    },
    {
      "lang": "fr",
      "title": "Apprentissage automatique",
      "url": "https://fr.wikipedia.org/wiki/Apprentissage_automatique"
    }
  ],
  "cached": false,
  "timestamp": "2025-11-02T17:09:14.866Z"
}

wikipedia://related/{title}

Lists related articles and see-also links for a Wikipedia article.

Resource Details:

• Purpose: Discover related content and connections
• Benefits: Content discovery and research expansion
• Use Cases: Research expansion, content recommendations, topic exploration

Response Format:

{
  "title": "Machine Learning",
  "seeAlsoLinks": [
    { "ns": 0, "title": "Data mining" },
    { "ns": 0, "title": "Pattern recognition" }
  ],
  "relatedArticles": [
    "Artificial intelligence",
    "Statistics",
    "Computer science"
  ],
  "categories": [
    { "ns": 14, "title": "Category:Machine learning" }
  ],
  "cached": false,
  "timestamp": "2025-11-02T17:09:14.866Z"
}

wikipedia://summary/{title}

Delivers cached summary and key facts for a Wikipedia article.

Resource Details:

• Purpose: Get article essence without full content
• Benefits: Quick information retrieval and content preview
• Use Cases: Content preview, fact-checking, quick reference, research planning

Response Format:

{
  "title": "Machine Learning",
  "summary": "Machine learning is a method of data analysis that automates analytical model building...",
  "keyFacts": {
    "Developed": "1950s",
    "Key figures": ["Arthur Samuel", "Tom Mitchell"],
    "Applications": ["Computer vision", "Natural language processing"]
  },
  "pageid": 233488,
  "lastModified": "2025-11-02T15:30:00Z",
  "wordCount": 1250,
  "cached": false,
  "timestamp": "2025-11-02T17:09:14.866Z"
}

Usage Examples

Once configured in your MCP client (like Cursor), you can use natural language to interact with Wikipedia:

• "Search for articles about artificial intelligence"
• "Get the Wikipedia page for Machine Learning"
• "Find Wikipedia articles near New York City"
• "Show me the summary of Albert Einstein"
• "What languages is the Python programming language article available in?"

Accessing Resources

Resources can be accessed directly using the MCP resources/read method:

{
  "jsonrpc": "2.0",
  "method": "resources/read",
  "params": {
    "uri": "wikipedia://summary/Albert Einstein"
  },
  "id": 1
}

Common Resource URIs:

• wikipedia://summary/{title} - Get article summary
• wikipedia://article/{title}/{lang} - Get full article content
• wikipedia://metadata/{title} - Get article metadata
• wikipedia://languages/{title} - Get available languages
• wikipedia://categories/{title} - Get category information
• wikipedia://related/{title} - Get related articles
• wikipedia://search/cache/{query} - Get cached search results

Testing

The server includes a comprehensive Jest test suite:

npm test

Test Coverage:

• Service instantiation and method availability
• Wikipedia service integration
• Type safety and error handling

License

MIT License - see LICENSE file for details.