fastmcp-example
Verified Safeby gauravsingh8026
Overview
This project demonstrates building an AI agent workflow using an MCP (Model Context Protocol) server, integrating LangChain for tool usage, and LangGraph for multi-step orchestration.
Installation
python server.pyEnvironment Variables
- TAVILY_API_KEY
- OPENAI_API_KEY
Security Notes
The `http_request` tool (available both server-side via MCP and client-side via custom tools) allows making arbitrary HTTP requests. If an AI agent has unconstrained access to this tool, it could potentially be prompted to perform Server-Side Request Forgery (SSRF) to internal networks, Denial of Service (DoS) attacks, or interact with sensitive external APIs. The code does not include explicit URL allow-listing or advanced input sanitization for these network operations. No `eval` or similar direct code execution vulnerabilities were found, and API keys are managed via environment variables.
Similar Servers
mcpstore
MCPStore acts as an orchestration layer for managing Microservice Context Protocol (MCP) services and adapting them as tools for AI frameworks like LangChain, AutoGen, and others.
mcp-servers
An MCP Server for robust web content fetching, anti-bot bypassing, intelligent caching, and LLM-powered information extraction from the open internet, designed for agent-building frameworks and MCP clients.
agents-mcp-usage
Demonstrates how to build a Model Context Protocol (MCP) server and integrate various agent frameworks.
ai-mcp-server-client-fastmcp
The server provides an MCP interface for an AI agent to access real-time stock market data via YFinance tools and respond to financial queries.