🤖 MultiAgentic Chart Generator

MultiAgent is an autonomous multi-agent AI system that collaborates to perform real-time research and generate charts — all via self-coordinated agents built on LangGraph & LangChain.

Turn your natural language queries into fully automated data analysis and visualization!

---

🧩 Multi-Agent Workflow Overview

Two specialized agents collaborate:

• Researcher Agent: Performs web-based research using the Tavily Search Tool to gather concise, structured data.
• Chart Generator Agent: Takes the research data and generates charts by executing Python code (e.g., matplotlib).

Agents take turns, reason independently, and collaborate to complete tasks end-to-end.

---

🚀 Core Features

Feature	Description
🧠 LangGraph	Enables stateful multi-agent coordination
🤖 LangChain React Agents	Each agent reasons & selects tools to use intelligently
🔍 Tavily Search Tool	Web research tool to fetch real-time data
💻 Python REPL Tool	Executes chart-generating Python code safely
🎯 Custom System Prompts	Precisely defines agent behavior and handover logic
🔁 Autonomous Transitions	Agents switch roles until FINAL ANSWER is found

---

🛠️ Tech Stack & Tools

Tool	Description	Icon
LangGraph	Graph-based multi-agent workflow engine
LangChain	Framework for LLM agents and tool use
Gemini 1.5 Flash	LLM backend for fast & intelligent decisions
Tavily Search Tool	Real-time web search API
Python REPL Tool	Executes chart code in real-time

---

💡 Example Query

Get the UK's GDP over the past 3 years and then generate a line chart. Once the chart is ready, end the process.

---

⚙️ Installation & Running

1. Clone the repository and navigate to the Agentic-2.0/langgraph directory:

   git clone <repository-url>
   cd Agentic-2.0/langgraph

2. Install the required dependencies:

   pip install -r requirements1.txt

3. Set up environment variables for API keys by creating a .env file in the root directory with:

   GROQ_API_KEY=your_groq_api_key_here
   GOOGLE_API_KEY=your_google_api_key_here
   TAVILY_API_KEY=your_tavily_api_key_here
   

4. Run the MulriAgent.ipynb notebook to start the multi-agent system.

---

📄 License

This project is licensed under the MIT License. See the LICENSE file for details.

---