Provides tools for searching Wikipedia and retrieving page content, enabling AI agents to access Wikipedia knowledge.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@wikipedia Search Wikipedia for the history of artificial intelligence"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
An AI-powered Wikipedia research agent demonstrating how FastAPI, FastMCP, LangChain, and LangGraph can work together to build a modular tool-using AI application.
The project exposes Wikipedia capabilities through an MCP (Model Context Protocol) server and allows a LangChain/LangGraph agent to automatically discover and use those tools.
┌──────────────────────┐
│ User │
│ Natural Language │
└──────────┬───────────┘
│
▼
┌──────────────────────┐
│ FastAPI │
│ Application API │
└──────────┬───────────┘
│
▼
┌──────────────────────┐
│ LangChain Agent │
│ LangGraph │
└──────────┬───────────┘
│
MCP Protocol
│
▼
┌──────────────────────┐
│ FastMCP │
│ MCP Server │
└──────────┬───────────┘
│
┌──────────┴───────────┐
▼ ▼
┌──────────────┐ ┌──────────────┐
│ Wikipedia │ │ Wikipedia │
│ Search │ │ Page │
└──────────────┘ └──────────────┘Related MCP server: MCP Wikipedia Server
🤖 LLM-powered Wikipedia research
🔌 Model Context Protocol (MCP) integration
⚡ FastAPI application layer
🧠 LangChain agent integration
🔄 LangGraph ReAct agent
🔎 Wikipedia search tool
📄 Wikipedia page retrieval tool
🔗 MCP tool discovery
📡 stdio-based MCP communication
🧩 Modular architecture that can easily support additional tools
Technology | Purpose |
Python | Core programming language |
FastAPI | API/application layer |
FastMCP | MCP server and tool implementation |
LangChain | LLM and tool integration |
LangGraph | Agent workflow |
Requests | Wikipedia API requests |
Wikipedia REST API | External knowledge source |
The FastMCP server exposes two tools.
search_wikipediaSearches Wikipedia for a given topic.
@mcp.tool
def search_wikipedia(query: str):
...Example:
Search Wikipedia for Albert EinsteinThe agent can automatically decide to call:
search_wikipedia("Albert Einstein")get_wikipedia_pageRetrieves the content of a specific Wikipedia page.
@mcp.tool
def get_wikipedia_page(title: str):
...Example:
Get the Wikipedia page for Artificial IntelligenceThe agent can call:
get_wikipedia_page("Artificial Intelligence")The MCP server runs using:
mcp.run(transport="stdio")The LangChain client connects to the MCP server:
client = MultiServerMCPClient(
{
"wikipedia": {
"command": "python",
"args": ["server.py"],
"transport": "stdio",
}
}
)The client then discovers the available MCP tools:
tools = await client.get_tools()These tools are passed to the LangGraph agent:
agent = create_react_agent(
model,
tools
)The LLM can then decide which tool to use based on the user's request.
git clone https://github.com/YOUR_USERNAME/langchain-mcp-wikipedia.gitcd langchain-mcp-wikipediaWindows:
python -m venv venv
venv\Scripts\activateLinux/macOS:
python3 -m venv venv
source venv/bin/activatepip install -r requirements.txtCreate a .env file:
OPENAI_API_KEY=your_openai_api_keyTraditional tool integration often tightly couples an LLM application with individual APIs.
MCP provides a standardized way to expose capabilities as tools.
LLM
│
▼
LangChain / LangGraph
│
▼
MCP Client
│
▼
MCP Server
│
├── Wikipedia
├── Search
├── Database
├── APIs
└── Custom ToolsThis separation makes tools reusable across different AI applications and agents.
This project is useful for learning:
Model Context Protocol (MCP)
FastMCP
MCP servers
MCP clients
stdio transport
LangChain tool integration
LangGraph agents
ReAct agents
FastAPI
External API integration
LLM tool calling
Modular AI agent architecture
This server cannot be deployed
Search Wikipedia, read summaries and full text, target sections, find nearby pages, list languages.
Cited, versioned knowledge for agents: retrieve sourced passages and propose owner-approved fixes.
Direct access to 60+ scraping and search tools. Extract structured data from Google (Search, Maps, Trends), Amazon, Airbnb, Social Media, and any web page directly into your AI agent.
Real-time Amazon, WIPO & PACER data for AI agents — 19 tools via the MCP protocol.