CaseCanopy

Overview

CaseCanopy is an AI-powered legal platform that bridges the justice gap by enabling cross-jurisdiction legal precedent discovery and outcome prediction. It provides equal access to legal insights through a modern, responsive web application.

Project Structure

• agentic-ai/: Python FastAPI service for legal document generation using AI.
• backend/: Go (Gin) backend for user, admin, file, and document management.
• frontend/: Next.js/React frontend for user interaction and legal research.
• RAG/: Python Flask server for retrieval-augmented generation (LangChain-based).
• mcp_server/: Model Context Protocol server for Claude Desktop integration, providing legal precedent search, document generation, and analysis tools.

Key Features

• AI-powered legal document generation (petitions, RTIs, complaints, etc.)
• Case law and legal precedent search
• Outcome prediction and legal insights
• User authentication and admin approval
• Document upload, management, and PDF generation
• Modern, responsive frontend UI
• Claude Desktop integration via MCP server for seamless AI legal assistance

Setup Instructions

1. Clone the repository:

git clone https://github.com/Arpit529Srivastava/Case_Canopy.git
cd Case_Canopy

2. Set up the AI Agent:

cd ai_agent
   # Follow instructions in ai_agent/README.md

3. Set up the Backend:

cd backend
go mod tidy
go run main.go
# Server runs on :8000, requires MongoDB running locally

4. Set up the Frontend:

cd frontend
npm install
npm run dev
# App runs on http://localhost:3000

5. Set up the RAG:

cd RAG
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python app.py
# Server runs on http://localhost:8000

# in other terminal tab run:
source venv/bin/activate
pip install -r requirements.txt
python analyzer.py

6. Set up the MCP Server (Claude Desktop Integration):

cd mcp_server
npm install

Configure Claude Desktop:

1. Add configuration to Claude Desktop config file:
   • Windows: %APPDATA%\Claude\claude_desktop_config.json
   • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
   • Linux: ~/.config/claude/claude_desktop_config.json

{
  "mcpServers": {
    "case-canopy-legal": {
      "command": "node",
      "args": ["C:/path/to/your/case-canopy/mcp_server/index.js"],
      "env": {
        "RAG_BASE_URL": "http://localhost:9000",
        "AI_AGENT_BASE_URL": "http://localhost:8001"
      }
    }
  }
}

2. Ensure RAG system (port 9000) and AI agent (port 8001) are running
3. Restart Claude Desktop

MCP Server Features:

• 🔍 Legal Precedent Search: Search Supreme Court cases and jurisprudence
• 📄 Document Generation: Auto-generate PIL, RTI, and Complaint documents
• ⚖️ Combined Analysis: Get legal analysis with document generation in one step

Environment Variables

1. agentic-ai/.env

OPENAI_API_KEY=your_openai_api_key_here

2. RAG/.env

OPENAI_API_KEY=your_openai_api_key_here
MODEL_NAME=gpt-4o-mini 
QDRANT_URL=your_link
QDRANT_API_KEY=your_qdrant_api_key_here

3. backend/.env

SMTP_HOST=smtp.gmail.com
SMTP_PORT=587
SMTP_USER=your_email
SMTP_PASS=generate_password_and paste_here
JWT_SECRET=your_token
GEMINI_API_KEY=your_api_key

4. frontend/.env.local

MONGODB_URI=you_uri
JWT_SECRET=secret_token

License

• This project is licensed under the MIT License.