MCP Wikipedia Server

import wikipedia import asyncio import sys from datetime import datetime from typing import Dict, Any from mcp.server.fastmcp import FastMCP mcp = FastMCP("WikipediaSearch") class WikipediaServer: """ Wikipedia Server class for testing and programmatic access. This class provides the same functionality as the MCP tools but in a class-based interface that's easier to test and use programmatically. """ def __init__(self): self.name = "WikipediaServer" self.version = "1.0.0" async def fetch_wikipedia_info(self, query: str) -> Dict[str, Any]: """ Search Wikipedia for a topic and return comprehensive information. Args: query: Search query string Returns: Dictionary containing success status, data, and metadata """ try: # Input validation if not query or not isinstance(query, str): return { "success": False, "error": "Query must be a non-empty string", "metadata": { "query": str(query) if query else "", "timestamp": datetime.now().isoformat() } } query = query.strip() if not query: return { "success": False, "error": "Query cannot be empty or whitespace only", "metadata": { "query": query, "timestamp": datetime.now().isoformat() } } # Search Wikipedia search_results = wikipedia.search(query, results=10) if not search_results: return { "success": False, "error": "No Wikipedia articles found for this query", "metadata": { "query": query, "timestamp": datetime.now().isoformat() } } # Get the best match best_match = search_results[0] page = wikipedia.page(best_match) return { "success": True, "data": { "title": page.title, "summary": page.summary, "url": page.url, "page_id": page.pageid, "length": len(page.content), "categories": page.categories[:10], # Limit to first 10 "links": page.links[:20], # Limit to first 20 "last_modified": datetime.now().isoformat() # Placeholder }, "metadata": { "query": query, "search_time_ms": 0, # Placeholder "timestamp": datetime.now().isoformat() } } except wikipedia.DisambiguationError as e: return { "success": False, "error": f"Multiple articles found. Please be more specific.", "error_type": "DisambiguationError", "suggestions": list(e.options[:10]), "metadata": { "query": query, "timestamp": datetime.now().isoformat() } } except wikipedia.PageError: return { "success": False, "error": "No Wikipedia page could be loaded for this query", "error_type": "PageError", "metadata": { "query": query, "timestamp": datetime.now().isoformat() } } except Exception as e: return { "success": False, "error": f"An error occurred: {str(e)}", "error_type": type(e).__name__, "metadata": { "query": query, "timestamp": datetime.now().isoformat() } } async def list_wikipedia_sections(self, topic: str) -> Dict[str, Any]: """ Return a list of section titles from the Wikipedia page of a given topic. Args: topic: Wikipedia article title or search query Returns: Dictionary containing success status, sections data, and metadata """ try: # Input validation if not topic or not isinstance(topic, str): return { "success": False, "error": "Topic must be a non-empty string", "metadata": { "topic": str(topic) if topic else "", "timestamp": datetime.now().isoformat() } } topic = topic.strip() if not topic: return { "success": False, "error": "Topic cannot be empty or whitespace only", "metadata": { "topic": topic, "timestamp": datetime.now().isoformat() } } # Get Wikipedia page page = wikipedia.page(topic) sections = page.sections # Structure sections with additional metadata structured_sections = [] for i, section in enumerate(sections): structured_sections.append({ "title": section, "level": 2, # Default level, could be enhanced "index": i + 1 }) return { "success": True, "data": { "article_title": page.title, "sections": structured_sections, "total_sections": len(sections) }, "metadata": { "topic": topic, "page_id": page.pageid, "timestamp": datetime.now().isoformat() } } except wikipedia.PageError: return { "success": False, "error": f"Wikipedia article not found for topic: {topic}", "error_type": "PageError", "metadata": { "topic": topic, "timestamp": datetime.now().isoformat() } } except Exception as e: return { "success": False, "error": f"An error occurred: {str(e)}", "error_type": type(e).__name__, "metadata": { "topic": topic, "timestamp": datetime.now().isoformat() } } async def get_section_content(self, topic: str, section_title: str) -> Dict[str, Any]: """ Return the content of a specific section in a Wikipedia article. Args: topic: Wikipedia article title or search query section_title: Name of the section to retrieve Returns: Dictionary containing success status, section content, and metadata """ try: # Input validation if not topic or not isinstance(topic, str): return { "success": False, "error": "Topic must be a non-empty string", "metadata": { "topic": str(topic) if topic else "", "section_title": str(section_title) if section_title else "", "timestamp": datetime.now().isoformat() } } if not section_title or not isinstance(section_title, str): return { "success": False, "error": "Section title must be a non-empty string", "metadata": { "topic": topic, "section_title": str(section_title) if section_title else "", "timestamp": datetime.now().isoformat() } } topic = topic.strip() section_title = section_title.strip() if not topic or not section_title: return { "success": False, "error": "Topic and section title cannot be empty or whitespace only", "metadata": { "topic": topic, "section_title": section_title, "timestamp": datetime.now().isoformat() } } # Get Wikipedia page page = wikipedia.page(topic) content = page.section(section_title) if content is None: # Try to find similar sections available_sections = page.sections return { "success": False, "error": f"Section '{section_title}' not found in article '{page.title}'", "error_type": "SectionNotFound", "suggestions": available_sections[:10], "metadata": { "topic": topic, "section_title": section_title, "page_id": page.pageid, "timestamp": datetime.now().isoformat() } } # Handle empty content if not content: content = "" return { "success": True, "data": { "article_title": page.title, "section_title": section_title, "content": content, "content_length": len(content), "subsections": [], # Could be enhanced to find subsections "section_level": 2 # Default level }, "metadata": { "topic": topic, "section_title": section_title, "page_id": page.pageid, "timestamp": datetime.now().isoformat() } } except wikipedia.PageError: return { "success": False, "error": f"Wikipedia article not found for topic: {topic}", "error_type": "PageError", "metadata": { "topic": topic, "section_title": section_title, "timestamp": datetime.now().isoformat() } } except Exception as e: return { "success": False, "error": f"An error occurred: {str(e)}", "error_type": type(e).__name__, "metadata": { "topic": topic, "section_title": section_title, "timestamp": datetime.now().isoformat() } } @mcp.tool() def fetch_wikipedia_info(query: str) -> dict: """ Search Wikipedia for a topic and return title, summary, and URL of the best match. """ try: search_results = wikipedia.search(query) if not search_results: return {"error": "No results found for your query."} best_match = search_results[0] page = wikipedia.page(best_match) return { "title": page.title, "summary": page.summary, "url": page.url } except wikipedia.DisambiguationError as e: return { "error": f"Ambiguous topic. Try one of these: {', '.join(e.options[:5])}" } except wikipedia.PageError: return { "error": "No Wikipedia page could be loaded for this query." } @mcp.tool() def list_wikipedia_sections(topic: str) -> dict: """ Return a list of section titles from the Wikipedia page of a given topic. """ try: page = wikipedia.page(topic) sections = page.sections return {"sections": sections} except Exception as e: return {"error": str(e)} @mcp.tool() def get_section_content(topic: str, section_title: str) -> dict: """ Return the content of a specific section in a Wikipedia article. """ try: page = wikipedia.page(topic) content = page.section(section_title) if content: return {"content": content} else: return {"error": f"Section '{section_title}' not found in article '{topic}'."} except Exception as e: return {"error": str(e)} if __name__ == "__main__": print("Starting MCP Wikipedia Server with multiple tools...") mcp.run(transport="stdio")