🚀 LangGraph Chatbot

AI-powered conversational agent built using LangGraph, FastAPI, and a multi-tool agent system capable of performing PDF search and mathematical calculations.

This project demonstrates how to build a production-ready agent workflow using LangGraph, integrate custom tools, and expose the assistant through a clean FastAPI backend.

---

🔥 Features

• LangGraph-based Agent Workflow
  Fully modular and deterministic agent pipeline built using LangGraph StateGraph.

• Multi-Tool Agent

  • 🔍 PDF Search Tool — Extracts & summarizes content from PDF documents
  • ➗ Math/Calculator Tool — Solves arithmetic expressions
  • 💬 LLM Default Tool — Used when no tool matches

• FastAPI Backend (/chat endpoint)

• Duplicate Response Handling

• Clean Code Structure

---

🧠 Architecture Overview

User Message
↓
LangGraph StateGraph
↓
Tool Router
├── PDF Search Tool
├── Calculator Tool
└── LLM Tool (default)
↓
Final Aggregated Response
↓
FastAPI Response → User

📁 Project Structure

LangGraph-Chatbot/
├── Chatbot/
│ └── chatbot.py # LangGraph workflow, tools, routing logic
├── chatbot_api.py # FastAPI API for chatbot
├── requirements.txt # Dependencies
├── pyproject.toml # Project metadata
├── README.md # Documentation
├── .gitignore # Ignored files
└── uv.lock # Dependency lock file

⚙️ Installation & Setup

1️⃣ Clone Repository

git clone https://github.com/Saurav-kumar077/LangGraph-Chatbot.git
cd LangGraph-Chatbot

2️⃣ Create Virtual Environment

python -m venv .venv

3️⃣ Activate Environment

Windows:

.\.venv\Scripts\Activate.ps1

4️⃣ Install Dependencies

pip install -r requirements.txt

🛠️ Tech Stack

• Python 3.12+
• LangGraph (modular agent workflows)
• FastAPI (API server)
• LangChain (tool integration)
• Pydantic (validation)
• Uvicorn (ASGI server)

👨‍💻 Author

Saurav Kumar
AI & Mobile Developer
GitHub: https://github.com/Saurav-kumar077