Legal AI Contract Intelligence Platform

Project developed for the Cyfuture AI Hackathon

---

🚀 Overview

Legal documents are notoriously lengthy, dense, and rich in specialized jargon, making comprehension a significant challenge. Our platform integrates a robust AI stack combining Gemini, LangChain, and DuckDuckGo to deliver end-to-end legal intelligence.

---

🏆 Key Features

• 🔐 Login & Access: Secure Google/OTP authentication.
• 📄 Contract Upload: Upload contracts (PDF/images) for instant analysis.
• 📊 Analysis & Red Flag Detection: OCR (Tesseract/Azure Vision) extracts text; LangChain + Gemini identify clauses and compliance issues.
• 💬 Contract Chat Interface: Chat with your document using Gemini-powered semantic chat (RAG), with contextual Q&A and clause referencing.
• 🔍 Similar Contract Search: DuckDuckGo scraping retrieves similar contracts; Gemini ranks them by similarity and success probability.
• 📥 PDF Export: Export chat transcripts and insights as polished PDFs for compliance, audit, and collaboration.
• 🧠 Gemini Prompted Chatbot: Domain-aware Gemini chatbot for general legal queries.
• Contextual Contract Benchmarking: One-click search for similar contracts or past agreements for negotiation insights.
• Semantic Document Chat: RAG-powered Q&A with full context—no keyword guessing.
• Downloadable Knowledge Artifacts: Export conversations with embedded context as polished PDFs.

---

🧑‍💻 Tech Stack

• Frontend: Next.js, React, Tailwind CSS
• Backend: Node.js, Express, MongoDB, Passport.js
• AI: Google Gemini, LangChain, DuckDuckGo
• OCR: Tesseract, PyMuPDF, Python integration
• PDF: pdfkit

---

🛠️ Installation & Setup

1. Clone the repository

git clone <repo-url>
cd legal_ai

2. Install dependencies

Frontend

cd client
npm install

Backend (Node.js)

cd ../server
npm install

Backend (Python OCR)

pip install -r requirements.txt

3. Install Tesseract OCR

• Windows: Download from here and add to PATH.
• macOS: brew install tesseract
• Linux: sudo apt-get install tesseract-ocr

4. Environment Variables

• Copy .env.example to .env in both client and server folders and fill in required keys (Google, Cloudinary, MongoDB, Gemini, etc).

5. Start the servers

# In /server
npm start
# In /client
npm run dev

---

📖 API Documentation (Backend)

Authentication

• POST /api/auth/send-otp — Send OTP to email
• POST /api/auth/verify-otp — Verify OTP, get JWT
• GET /api/auth/me — Get user info (JWT required)
• GET /api/auth/google — Google OAuth login

OCR & Document Upload

• POST /api/ocr/upload-single — Upload and OCR a single file
• POST /api/ocr/upload-multiple — Upload and OCR multiple files
• GET /api/ocr/result/:fileId — Get OCR result by fileId
• GET /api/ocr/chunks/:fileId?page=1&limit=10 — Paginated text chunks
• GET /api/ocr/history — User's OCR upload history

Legal Analysis

• POST /api/legal/analyze/:fileId — Analyze contract for legal structure, risks, compliance, red flags
• POST /api/legal/summary/:fileId — Executive summary
• POST /api/legal/entities/:fileId — Extract key entities/terms

Chatbot

• POST /api/legal/initialize — Start chat session
• POST /api/legal/ask — Chat with legal AI (contextual)
• GET /api/legal/history — Get chat history
• POST /api/legal/clear — Clear chat history

Web Search & Benchmarking

• POST /api/websearch/search-contracts — Find and rank similar contracts using DuckDuckGo + Gemini

PDF Generation

• POST /api/document/generate — Generate PDF summary of consultation

---

🧩 Python OCR Service

• Script: server/ocr.py
• Install dependencies:

  pip install -r requirements.txt

• Run as part of backend Node.js service (auto-invoked)
• Libraries:
  • pytesseract, Pillow, PyMuPDF, langchain-text-splitters, langchain-core

---

🎯 Target Customers

• Legal teams & law firms (NDAs, vendor agreements, M&A)
• In-house corporate counsel (compliance, onboarding)

📈 Market Insights

• 2024: US $31.6 B → 2032: US $63.6 B (CAGR 9.4%)
• North America ~50% share; APAC fastest-growing
• Legal AI/Contract AI: strong VC interest (Harvey, Ivo, etc)

📣 Marketing & Outreach

• Webinars, demos, case studies, legal tech conferences
• Freemium model: limited analysis, premium for PDF/benchmarking

---

🤝 Acknowledgements

• Built for the Cyfuture AI Hackathon
• Powered by Google Gemini, LangChain, DuckDuckGo, Tesseract, PyMuPDF

---