OpenDAIL

OpenDAIL is a web application for analyzing and exploring AI litigation cases from the DAIL database. It combines a FastAPI backend with a React frontend to provide AI-powered querying, interactive charts, and comprehensive case management.

Features

AI-Powered Query Interface

• Natural language to SQL conversion using Ollama LLM(Uses gpt oss 120B which is open source)
• Ask questions about case data in plain English
• Automatic SQL query generation and execution
• Results transformed into chart-ready formats

Interactive Dashboard

• Real-time charts and visualizations using Recharts
• Temporal analysis of case data
• Multiple chart types: bar, line, pie, and donut charts
• Summary statistics and lifecycle analysis

AI Chat Interface

• Conversational interface for querying case data
• Real-time responses with streaming indicators
• Inline chart rendering from query results
• Natural language interaction with the database

Case Management

• Browse all cases in a searchable table view
• Advanced filtering by multiple columns
• Date range filtering for temporal queries
• Collapsible filter panel for easy access

Detailed Case View

• Comprehensive case information display
• Case overview and significance summary
• Metadata: jurisdiction, dates, status, researcher
• Classifications: areas of application, issues, causes of action, algorithms
• Secondary sources with external links

Diagrams & Statistics

• Generate filtered statistics and visualizations
• Apply multiple filters before processing
• Visual charts based on filtered data
• Empty state handling for no-match queries

Tech Stack

Backend

• FastAPI - Modern Python web framework
• SQLite - Lightweight database for case storage
• Ollama - Local LLM for natural language processing

Frontend

• React Router - Full-stack React framework
• Recharts - Composable charting library
• Tailwind CSS - Utility-first CSS framework
• Shadcn/UI - Reusable component library
• MUI - Material UI component library

Getting Started

Prerequisites

• Python 3.10+
• Node.js 18+
• Ollama server running locally

Backend Setup

cd backend
pip install -r requirements.txt
python main.py

The backend runs on http://localhost:5174

Frontend Setup

cd frontend
npm install
npm run dev

The frontend runs on http://localhost:5173

API Endpoints

Dashboard

• GET /api/dashboard - Get chart data and temporal analysis
• GET /api/cases - Get all cases with optional filters
• GET /api/case/{case_id} - Get detailed case information
• GET /api/diagrams - Get filtered diagram data

Agents

• POST /api/agents/process - Process natural language query
• GET /api/agents/crews - Get available agent crews

Project Structure

DefinitelyNotAI/
├── backend/
│   ├── agents/
│   │   └── tools/
│   │       └── sql_tools.py    # SQL execution utilities
│   ├── routes/
│   │   ├── agents.py           # AI agent endpoints
│   │   └── dashboard.py       # Dashboard endpoints
│   ├── utils/
│   │   └── dashboard_utils/
│   │       └── analyse_data.py # Data analysis utilities
│   └── main.py                 # FastAPI application
└── frontend/
    ├── app/
    │   ├── api/               # API client functions
    │   ├── components/       # React components
    │   │   ├── ai-elements/  # AI chat components
    │   │   ├── charts/       # Chart components
    │   │   ├── home/         # Home page components
    │   │   └── ui/           # UI components
    │   ├── routes/           # Page routes
    │   └── types/            # TypeScript types
    └── package.json

Usage

1. Home Dashboard - View charts and ask AI questions about the data
2. All Cases - Browse and filter through all cases
3. Case Details - Click on any case to see full details
4. Diagrams - Apply filters and generate custom visualizations
5. Chat - Have a conversation with the AI about case data

Use of AI

We used AI coding tools including OpenCode and Cursor alongside writing code ourselves, and leveraged Claude, Gemini, and ChatGPT for research throughout the process.

Demo link - Demo video OpenDAIL