📝 FAQ Bot (CLI) with LangGraph & Fuzzy Matching

A simple command-line FAQ bot built using LangGraph and fuzzy matching. It allows users to interact with a predefined set of FAQs in a conversational manner. The bot uses fuzzy logic to match user input even with slight spelling variations or case mismatches.

---

🚀 Features

• Conversational CLI interaction 🤖
• Fuzzy matching of questions 🔍
• Loads questions from an editable JSON file 📁
• Custom bot name using environment variables 🔐
• Lightweight, fast, and offline-capable 🏃‍♂️

---

🛠️ Tech Stack

• Python
• LangGraph
• FuzzyWuzzy
• python-dotenv
• json

---

📁 FAQ Data Format

FAQs are stored in a separate JSON file (faq_data.json) like this:

[
  {
    "question": "What is LangGraph?",
    "answer": "LangGraph is a framework for building stateful AI agents using graphs."
  },
  {
    "question": "How do I run this bot?",
    "answer": "Just activate your virtual environment and run python faq_bot.py."
  }
]

---

🔐 Environment Setup

Store your database credentials in a .env file:

BOT_NAME=LangFAQBot

Important: The .env file is excluded via .gitignore to keep credentials secure.

📦 Installation

Clone this repository

git clone https://github.com/CodeSaadUmair/faq-bot-cli.git
cd faq-bot-cli

Create virtual environment

python -m venv envo
source envo/bin/activate  

or on Windows

python -m venv envo
envo\Scripts\activate 

Install dependencies

pip install -r requirements.txt

Run the app

python faq_bot.py

📌 Sample Questions

• how do I update my profile information?
• update my profil
• do you ship internationally?
• how can I reset my password?
• reset my password

✅ To-Do / Future Improvements

• Add persistent logging of user queries 📜
• Integrate OpenAI for fallback answers 🤯
• Add a web UI using Flask or Streamlit 🌐
• FAQ categories or topic filters 🗂️