Sentilex AI Advocate

Autonomous Legal Support & Forensic Evidence Management System

"Bridging the Gap Between Trauma and Justice"

🌐 Visit Live Site

---

🏛️ Vision

Sentilex is a pioneering multi-agent AI system designed to democratize legal access in Sri Lanka. It serves as a first-response legal advocate, conducting empathetic, trauma-aware interviews with victims, securing forensic-grade evidence via cryptographic hashing, and instantly generating court-admissible documentation compliant with the Computer Crimes Act No. 24 of 2007 and Online Safety Act No. 9 of 2024.

---

🚀 Key Features

🕊️ Trauma-Aware Legal Interviewing

AI-powered empathetic conversations guide victims through incident reporting. The system uses intent-based routing to adapt responses for emergencies, legal guidance, and evidence collection without requiring prior legal knowledge.

⚖️ Intelligent Legal Analysis (RAG)

Retrieves and cites relevant provisions from Sri Lankan Law. Uses a Model Context Protocol (MCP) with full audit trails to prevent hallucinations, ensuring advice is legally sound and contextually accurate.

🔐 Forensic-Grade Evidence Vault

Processes uploaded files with SHA-256 cryptographic hashing and metadata tracking (timestamps, S3 keys) to create a tamper-proof, court-admissible chain of custody.

📄 Professional Document Generation

Automatically generates submission-ready Police Statements and CERT Technical Reports, complete with structured incident narratives and evidence inventories.

🛡️ Verified Lawyer Directory

A searchable database of verified legal professionals. Includes a rigid verification workflow where lawyers upload NICs and Attorney Certificates for admin validation.

---

🛠️ Technology Stack

Frontend

Backend & AI

Infrastructure & DevOps

---

🏗️ System Architecture

Sentilex employs a secure, containerized microservices architecture:

1. Entry Point: Traffic is routed through Cloudflare to an Nginx Reverse Proxy.
2. Application Layer: Dockerized containers for Frontend (Preact) and Backend (FastAPI).
3. Authentication: Robust IAM system handling Users, Lawyers, and Admins with JWT, OAuth (Google), and MFA (TOTP).
4. Data Layer: PostgreSQL handles relational data (Users, Incidents), while AWS S3 stores encrypted evidence.
5. AI Engine: A multi-agent system powered by LangGraph handles intent classification, legal reasoning, and safety validation.
6. Real-Time Updates: WebSocket server pushes notifications (e.g., verification status) instantly to clients.

---

⚡ Quick Start with Docker

Run the entire application with a single command.

Prerequisites

• Docker & Docker Compose

🚀 Launch

1. Clone the repository

   git clone https://github.com/Smile-Labs-Inc/sentilex-ai-advocate.git
   cd sentilex-ai-advocate

2. Environment Setup Create a .env file from the example.

   cp .env.example .env

3. Start Services

   docker compose up -d --build

The application will be available at:

• Frontend: http://localhost:8080
• API Docs: http://localhost:8001/docs

---

🤝 Contributing

We welcome contributions! Please see CONTRIBUTING.md for details on how to submit pull requests, report issues, and suggest improvements.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

---

_{Built with ❤️ by Smile Labs Inc. for a Safer Digital Sri Lanka.}