LawGeeks-Pro ⚖️

Demystifying Legal Jargon with AI-Powered Intelligence

An intelligent paralegal assistant that simplifies complex legal documents using Retrieval-Augmented Generation (RAG) and advanced document analysis. Built to bridge the gap between technical legal language and everyday understanding.

---

🚀 Key Features

• 📄 Multi-Format Support: Seamlessly analyze .pdf, .docx, images (.png, .jpg), or direct text input.
• 🔍 Automated Insights: Generates structured summaries, extracts critical dates, identifies financial terms, and highlights key clauses.
• 🛡️ Vigilance Score: A dynamic risk assessment (1–100) that flags unfair clauses and potential legal traps.
• 💬 RAG-Powered Q&A: Ask context-aware questions grounded in your document and a pre-loaded knowledge base of Indian Law.
• 📊 Professional Reporting: Export your analysis and insights into a clean, downloadable PDF report.
• 🌐 Accessibility: Integrated text-to-speech engine and translation support for major Indian languages.

---

🏗️ System Architecture & Workflow

LawGeeks-Pro utilizes a three-tier architecture combining a modern frontend, an asynchronous FastAPI backend, and a RAG (Retrieval-Augmented Generation) intelligence layer.

Workflow Diagram (Simplified)

graph TD
    User([User]) -->|Uploads Doc| FE[Web Frontend]
    User -->|Asks Question| FE
    FE -->|API Request| BE[FastAPI Backend]
    
    subgraph "AI Core"
        BE -->|Analysis| Gemini[Google Gemini Pro]
        BE -->|Retrieval| RAG[RAG Service]
        RAG -->|Semantic Query| VectorDB[(Chroma Vector DB)]
        VectorDB --- Search[Knowledge Base Search]
    end

    Gemini --> BE
    BE --> FE

Loading

For a deep-dive into the technical implementation, please see our Full System Architecture & Detailed Workflows document.

---

---

📂 File Structure

LawGeeks-Pro/
├── api/                    # Backend API Logic
│   ├── core/               # AI & RAG service implementations
│   ├── models/             # Pydantic data schemas
│   └── index.py            # Main FastAPI application
├── knowledge_base/         # PDF source documents for Indian Law
├── public/                 # Web Frontend assets
│   ├── assets/             # Images and icons
│   ├── css/                # Tailwind & Custom styles
│   ├── js/                 # Logic for analysis & chat
│   └── home.html           # Landing page
├── scripts/                # Utility & Setup scripts
│   ├── ingest.py           # Vector database builder
│   └── .env                # Environment configurations
├── vector_db/              # Persisted ChromaDB data
├── config/                 # Static configurations
├── tests/                  # Unit and integration tests
├── requirements.txt        # Python dependencies
└── vercel.json             # Deployment configuration

---

💻 Technolgoy Stack

• Backend: FastAPI, Uvicorn
• AI/LLM: Google Gemini (gemini-pro-latest)
• RAG Framework: LangChain, ChromaDB
• Embeddings: Google Generative AI Embeddings
• Frontend: HTML5, TailwindCSS, Vanilla JavaScript
• PDF/Docs: pypdf, python-docx, tesseract-ocr

---

🏁 Getting Started

1. Prerequisites

• Python 3.10+
• Google Gemini API Key (Get it here)

2. Installation

# Clone the repository
git clone https://github.com/SIBAM890/LawGeeks-Pro.git
cd LawGeeks-Pro

# Setup virtual environment
python -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

3. Configuration

1. Navigate to scripts/ and create a .env file:

   GOOGLE_API_KEY="your_api_key_here"

2. Place your reference legal PDFs in the knowledge_base/ folder.

4. Build Vector DB & Run

# Ingest the knowledge base
cd scripts
python ingest.py

# Start the server
cd ..
uvicorn api.index:app --reload

---

⚖️ Disclaimer

LawGeeks-Pro provides informational analysis only and should not be considered legal advice. Always consult a licensed legal professional for official guidance.

---

Proudly built for the Generative AI for Demystifying Legal Documents challenge.