AI_Agent_with_MCP

zayedrais/AI_Agent_with_MCP

3.3

If you are the rightful owner of AI_Agent_with_MCP and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

This project implements a Model Context Protocol (MCP) server using Google's Agent Development Kit (ADK) for building intelligent agents and tools.

The MCP Server with Google ADK Project is designed to facilitate the development and deployment of intelligent agents that can handle a variety of tasks through a coordinated approach. The server is built using Google's Agent Development Kit (ADK) and features a modular architecture where specialized agents collaborate under the guidance of a coordinator agent. This setup allows for efficient handling of requests, whether they involve data analysis, web searches, or code generation. The server is implemented using FastMCP, ensuring high performance and scalability. It supports multiple input data formats and provides outputs in the form of visualizations, reports, and generated code. The project is structured to allow easy extension, enabling developers to add new agents and tools as needed. With a focus on leveraging large language models (LLM) for intelligent routing and processing, the MCP server is a robust solution for automating complex workflows.

Features

  • LLM-based intelligent request routing through a coordinator agent.
  • Specialized agents for data analysis, web searches, and code generation.
  • Support for multiple data formats including CSV and Excel.
  • Extensible architecture for adding new agents and tools.
  • High-performance server implementation using FastMCP.

Usages

usage with local stdio

python server.py

usage with remote sse

mcp.run(transport='sse', host="0.0.0.0", port=8080)

usage with remote streamable http

paths:
  /mcp:
    post:
      x-ms-agentic-protocol: mcp-streamable-1.0

usage with fastmcp

from mcp.server import FastMCP
app = FastMCP('demo')
@app.tool()
async def query(): ...

Tools

  1. ask

    Routes the user's query to the most appropriate agent.

  2. search

    Searches the web for information.

  3. analyze_data

    Analyzes data files and generates reports with visualizations.

  4. generate_code

    Generates code based on natural language descriptions.