alekhakumarswain/AI-SECURITY-LABS
An advanced, interactive educational platform focused on AI system vulnerabilities, attack vectors, and offensive security methodologies. [Prompt Injection, Model Evasion, Data Poisoning, Agent Hijacking]
AI Security Labs
An advanced, interactive educational platform focused on AI system vulnerabilities, attack vectors, and offensive security methodologies.
[Adversarial Prompt Injection, Data Poisoning, Model Stealing, Agent Hijacking, etc.]
Course Overview | Hands-on Labs | GitHub Repository
๐ Overview
AI Security Labs is a comprehensive learning platform designed to teach the fundamentals and advanced concepts of AI security. The application is built using a modern Python FastAPI backend, rendering beautiful Vanilla CSS templates through Jinja2. All course data is modularly structured using YAML files, making it incredibly easy to expand the curriculum.
โจ Features
- Extensive Course Curriculum: Dive deep into topics like:
- Agent & Autonomous System Attacks (Tool Injection, SSRF, Excessive Agency)
- Adversarial Attacks & Prompt Injection
- Data Poisoning & Supply Chain Compromise
- Model Evasion & Model Stealing
- AI System Quality Testing & Red Teaming
- Modular Architecture: Courses are dynamically generated from rich YAML definitions. Adding a new module is as simple as dropping a new
.yamlfile. - Interactive UI/UX: Sleek, dark-mode focused aesthetic built specifically for security enthusiasts.
- Server-Side Rendering Context: Optimized for blazing-fast content delivery.
- Vercel Ready: Pre-configured
vercel.jsonfor immediate edge deployment.
๐ ๏ธ Tech Stack
- Backend framework: FastAPI
- Templating Engine: Jinja2
- Data Serialization: PyYAML
- Server: Uvicorn
- Frontend: HTML5, Vanilla CSS3 (Custom Design System)
๐ฆ Getting Started Locally
Prerequisites
Ensure you have Python 3.9+ installed on your system.
Installation
-
Clone this repository:
git clone https://github.com/alekhakumarswain/AI-SECURITY-LABS.git cd AI-SECURITY-LABS -
Install the required dependencies:
pip install -r requirements.txt
-
Run the development server:
python app.py
-
Open your browser and navigate to:
http://127.0.0.1:8000
๐ Deployment (Vercel)
This application is fully prepared for serverless deployment on Vercel. You can use the "Deploy with Vercel" button at the top of this documentation to instantly clone and deploy the app!
If you prefer using the Vercel CLI:
vercel๐ Project Structure
.
โโโ app.py # Main FastAPI application
โโโ requirements.txt # Python dependencies
โโโ vercel.json # Vercel deployment configuration
โโโ details/ # Course Data (YAML files)
โ โโโ adversarial_attacks.yaml
โ โโโ ai_agent_attacks.yaml
โ โโโ data_poisoning.yaml
โ โโโ ...
โโโ templates/ # Jinja2 HTML Templates
โ โโโ index.html # Landing Page
โ โโโ learn.html # Course Curriculum Page
โ โโโ labs.html # Hands-On Lab Terminal UI
โโโ static/ # Static Assets (CSS, Fonts, Images)
โโโ images/
โโโ fonts/
โ๏ธ Disclaimer
The information provided within AI Security Labs is for educational and research purposes only. Do not use these techniques to attack systems you do not have explicit permission to test.