Agentic Ai Mcp

What is Agentic Ai Mcp

Agentic-AI-MCP is an Agentic AI personal assistant built on LangGraph that integrates web research, database operations, and multimedia analysis into a single conversational interface. It utilizes a modular MCP architecture to intelligently route requests and maintain context throughout discussions.

Use cases

Use cases include retrieving information from the web, summarizing YouTube videos, performing database operations through natural language commands, and providing real-time weather updates.

How to use

To use Agentic-AI-MCP, clone the repository, set up a virtual environment, install the required dependencies, and configure your API keys. Start the weather server and the main agent in separate terminals, then interact with the application through the chat interface by entering your queries and commands.

Key features

Key features include YouTube video transcripts extraction, web research using Tavily API, database operations with SQLlite, real-time weather information, and conversation memory to maintain context during interactions.

Where to use

Agentic-AI-MCP can be used in various fields such as personal assistance, customer support, educational tools, and any application requiring intelligent conversational interfaces that integrate multiple data sources.

Agentic-AI-MCP

an Agentic AI personal assistant built on LangGraph that connectsvia MCP to the tools like web research, database operations, and multimedia analysis in one conversational interface. Uses modular MCP architecture to intelligently route requests and maintain context throughout discussions.

🚀 Features

• YouTube Video Transcipts - Extract and chat with the content of youtube videos
• Web Research - Search and retrieve information from the web using Tavily API
• Database Operations with SQLlite - Create tables, store data, and query information with natural language
• Weather Information - Get real-time weather data for any location
• Conversation Memory - Maintains context throughout interactions

📋 Prerequisites

• Python 3.9+
• An OpenAI API key
• A Tavily API key (for web search functionality)

🔧 Installation

1. Clone the repository:

git clone  https://github.com/Shivakumar980/Agentic-AI-MCP
cd langgraph-mcp

2. Create and activate a virtual environment:

python -m venv venv
# On Windows
venv\Scripts\activate
# On macOS/Linux
source venv/bin/activate

3. Install dependencies:

pip install -r requirements.txt

4. Create a .env file with your API keys:

OPENAI_API_KEY=your_openai_api_key
TAVILY_API_KEY=your_tavily_api_key

💻 Usage

Starting the Application

1. Start the weather server in one terminal:

python servers/weather.py

2. In another terminal, start the main agent:

python agent.py

3. Enter your queries and commands in the interactive chat interface.

Example Commands

Here are some examples of what you can do with KnowledgeForge:

Web Search

Tell me about Aravind Srinivasan

YouTube Transcripts

Summarize this YouTube video:https://www.youtube.com/watch?v=SP7Ua8FKZN4

Database Operations

Create a table called books with columns id, title, author, and genre

Add a book with title "Dune", author "Frank Herbert", and genre "Science Fiction"

Query the books table and show me all science fiction books

Update the book Dune to set its genre to "Sci-Fi Classic"

Weather Information

What's the weather like in San Francisco today?

🏗️ Architecture

The agentic application uses the Model Context Protocol (MCP) to connect a LangGraph agent with several specialized servers:

1. Tavily Server - Handles web search requests
2. YouTube Transcript Server - Extracts and processes video content
3. Weather Server - Provides weather data
4. Database Server - Manages SQLite database operations

The central agent routes requests to the appropriate servers and maintains conversation context.

📁 Project Structure

langgraph-mcp/
├── agent.py                 # Main agent with chat interface
├── requirements.txt         # Project dependencies
├── agent_database.db        # SQLite database file (created automatically)
├── .env                     # Environment variables for API keys
├── README.md                # This documentation
└── servers/                 # Directory containing MCP servers
    ├── tavily.py            # Web search server
    ├── yt_transcript.py     # YouTube transcript server
    ├── weather.py           # Weather information server
    └── database.py          # SQLite database server

🙏 Acknowledgements

• Built with LangGraph and Model Context Protocol
• Weather data provided by Open-Meteo
• Web search powered by Tavily