Finance AI Bot

The Finance AI Bot is an intelligent financial research assistant that provides real-time insights, analysis, and full-length reports on financial markets, stocks, and economic trends. It uses Groq's LLM API and direct web search (via DuckDuckGo) to deliver accurate, source-cited answers.

---

✨ Features

• Quick Answer Mode — Get a concise, bullet-point answer to any finance question in seconds, with live web sources.
• Deep Research Mode — Generate a comprehensive, 1000+ word Financial Report with executive summary, analysis, future outlook, and cited sources.
• Source Credibility Scores — Every response includes a color-coded confidence score (High / Medium / Low) based on the credibility of the sources used.
• XSS-Safe Output — All AI-generated HTML is sanitized with DOMPurify before rendering.
• Rate Limiting — API endpoints are protected (10 req/min for query, 5 req/min for research).
• Dockerized — Full Docker Compose setup for one-command deployment.

---

🛠️ Technologies Used

Category	Technologies
Backend	Python 3.11, Flask, Groq Python SDK, ddgs
Frontend	React 18, Axios, DOMPurify, Nginx
AI Model	Groq — llama-3.3-70b-versatile
Search	DuckDuckGo (via ddgs)
Infra	Docker, Docker Compose

---

🚀 Getting Started

Prerequisites

• Node.js 18+ and npm
• Python 3.11+ and pip
• A free Groq API key from console.groq.com

---

Local Development (without Docker)

1. Clone and configure

git clone https://github.com/Kshitij-KS/Finance-AI-Agent.git
cd Finance-AI-Agent/finbot

Copy the example env file and add your Groq API key:

cp .env.example .env
# Edit .env and set: groq_api_key=your_key_here

2. Start the Backend

cd Backend
pip install -r requirements.txt
python FinanceAI.py
# Server runs on http://localhost:5000

3. Start the Frontend

cd ../Frontend
cp .env.example .env.local   # already has REACT_APP_BACKEND_URL=http://localhost:5000
npm install
npm start
# App runs on http://localhost:3000

---

Docker Deployment (one command)

cd finbot
# Add your Groq key to .env first
docker compose up --build
# Frontend → http://localhost:80
# Backend  → http://localhost:5000

---

🌐 Free Cloud Deployment

Service	Platform	Notes
Backend	Render.com	Free tier, cold starts after idle
Frontend	Vercel.com	Free tier, always-on CDN

Steps:

1. Push Backend/ to GitHub → Deploy on Render as Python Web Service
   • Build command: pip install -r requirements.txt
   • Start command: gunicorn -w 2 -b 0.0.0.0:$PORT --timeout 120 FinanceAI:app
   • Env var: groq_api_key = <your key> and FRONTEND_URL = https://your-app.vercel.app
2. Push Frontend/ to GitHub → Deploy on Vercel (auto-detects CRA)
   • Env var: REACT_APP_BACKEND_URL = https://your-backend.onrender.com

---

📁 Project Structure

finbot/
├── Backend/
│   ├── FinanceAI.py          # Flask API — search + LLM logic
│   ├── requirements.txt      # Python dependencies
│   └── Dockerfile
├── Frontend/
│   ├── src/
│   │   ├── App.js            # Main React component
│   │   └── App.css           # All styles
│   ├── .env.example
│   └── Dockerfile            # Multi-stage Nginx build
├── compose.yaml              # Docker Compose
└── .env                      # API keys (gitignored)

---

⚙️ Environment Variables

Variable	Location	Description
groq_api_key	Backend .env	Your Groq API key
FRONTEND_URL	Backend .env	CORS allowed origin (production)
REACT_APP_BACKEND_URL	Frontend .env.local	Backend URL for API calls
FLASK_DEBUG	Backend .env	Set true for dev debug mode

---

🔒 Security Notes

• API keys are loaded from .env and never exposed to the frontend.
• CORS is restricted to localhost:3000 (dev) and your configured FRONTEND_URL (prod).
• All HTML rendered in the browser is sanitized with DOMPurify.
• Rate limiting prevents API abuse (10 req/min quick, 5 req/min research).

---

🙏 Acknowledgments

• Groq — Ultra-fast LLM inference
• DuckDuckGo (ddgs) — Privacy-respecting web search
• React — Frontend framework

---

📬 Contact

Email: kshitijsingh2727@gmail.com
GitHub: Kshitij-KS
Project: Finance-AI-Agent