SentinelIQ

Inspired by spy and detective movies, SentinelIQ is an AI-powered forensics analysis platform for immersive crime scene investigation. Users can upload evidence files, and the tool performs a comprehensive analysis using advanced machine learning techniques. It combines a modern React frontend with a Python FastAPI backend to provide real-time forensic analysis across multiple evidence types.

It is a Claude Desktop MCP server that can help to analyze crime scenes and get forensic reports from evidences.

Key Features

1. Crime Scene Analysis - Uses Ollama llava for detailed visual analysis
2. Fingerprint Analysis - Predicts blood group from fingerprint images using TensorFlow
3. Audio Transcription - Transcribes audio evidence using Faster Whisper
4. Suspect Identification - Matches suspects against a global database
5. Comprehensive Reporting - Generates integrated forensic reports combining all analyses

Technology Stack

Frontend

• React 19.2.4
• TypeScript 5.9.3
• Vite 8.0.1 (build tool)
• TailwindCSS 4.2.2
• Framer Motion 12.38.0 (animations)
• Lucide React 0.577.0 (icons)
• ESLint 9.39.4

Backend

• FastAPI (web framework)
• Uvicorn (ASGI server)
• TensorFlow (neural networks)
• PyTorch & Torchvision (deep learning)
• Faster Whisper (audio transcription)
• Transformers (LLM models)
• Scikit-learn & SciPy (ML utilities)
• Pillow (image processing)

Project Structure

ai-forensics-react/
├── src/                          # React frontend source
│   ├── main.tsx                  # Entry point
│   ├── App.tsx                   # Main app component with state management
│   ├── App.css
│   ├── index.css
│   ├── components/
│   │   ├── HeroSection.tsx       # Upload interface
│   │   ├── AnalyzingSection.tsx  # Live analysis progress
│   │   └── ReportSection.tsx     # Results display
│   └── assets/
├── backend/                       # Python backend
│   ├── app.py                    # FastAPI server (port 5500)
│   ├── image-a.py                # Crime scene & report analysis
│   ├── fingerprint.py            # Blood group prediction
│   ├── audio_agent.py            # Audio transcription
│   ├── suspect-identifying.py    # Suspect matching
│   ├── test_user_code.py
│   ├── requirements.txt           # Python dependencies
│   ├── fingerprint_bloodgroup_classifier_attention.h5  # Trained model
│   └── venv/                     # Virtual environment (to be created)
├── public/                        # Static assets
├── package.json                   # Frontend dependencies
├── vite.config.ts               # Vite configuration
├── tsconfig.json                # TypeScript configuration
└── README.md

Setup & Installation

Prerequisites

• Node.js 16+ and npm
• Python 3.8+
• Git

Frontend Setup

1. Navigate to the project root:

   cd ai-forensics-react

2. Install dependencies:

   npm install

Backend Setup

1. Navigate to backend directory:

   cd backend

2. Create a Python virtual environment:

   python -m venv venv

3. Activate the virtual environment:

   Windows (PowerShell):

   .\\venv\\Scripts\\Activate.ps1

   Windows (CMD):

   .\\venv\\Scripts\\activate.bat

   macOS/Linux:

   source venv/bin/activate

4. Install Python dependencies:

   pip install -r requirements.txt

Running the Project

Development Mode

Terminal 1 - Start the FastAPI backend:

cd backend
python app.py

(Backend runs on http://localhost:5500)

Terminal 2 - Start the React frontend:

npm run dev

(Frontend runs on http://localhost:5173)

3. Access the application:
   • Open http://localhost:5173 in your browser

Available npm Scripts

npm run dev      # Start Vite development server
npm run build    # Build for production (TypeScript + Vite)
npm run lint     # Run ESLint
npm run preview  # Preview production build

Claude Desktop Integration (MCP Server)

This project integrates with Claude Desktop via an MCP (Model Context Protocol) server to perform advanced AI-powered forensic analysis.

Prerequisites:

• Claude Desktop application must be installed.

Configuration:

1. Open the Claude Desktop application's configuration file (e.g., settings.json).

2. Locate the section for MCP server configurations.

3. Add or modify the entry to point to the project's root directory. This allows Claude to access the project's context for analysis.

   Example JSON configuration:

   {
     "mcp.server.paths": [
       "C:\\binary-frontend\\Binary2\\ai-forensics-react"
     ]
   }

   Note: The exact JSON key might differ. Please refer to the Claude Desktop documentation for the correct key.

4. Save the configuration file and restart Claude Desktop.

Once configured, you can leverage Claude's capabilities to interact with and analyze the project's data.