AI Fundamentals

Understand how LLMs work before you break them

Security Modules

Attack real LLM systems and learn how to mitigate the risks

Agentic AI

When AI systems act autonomously

Free Learning Path

Learn AI Security & Prompt Injection

More about this

A free, 9-module course covering LLM security from fundamentals to advanced attacks. Start with how LLMs work, then test real AI systems, bypass preconfigured defenses, and learn the mitigations in hands-on labs. No prior security experience required. Aligned with the OWASP Top 10 for LLM Applications and the OWASP Top 10 for Agentic Applications.

How LLMs Actually Work

Learn how large language models process tokens, context windows, and text generation - the foundation for understanding prompt injection attacks

12 minutes

System Prompts & the Context Window

How developers instruct LLMs with system prompts, why they're fragile, and how prompt injection exploits this fundamental weakness

15 minutes

RAG: When LLMs Read External Data

How Retrieval-Augmented Generation works, where trust boundaries break, and why RAG poisoning is a critical LLM security risk

12 minutes

Tools & Function Calling

How LLMs invoke external tools and APIs, and why tool exploitation and excessive agency are top LLM security risks

14 minutes

The Bare LLM

Direct prompt injection against unprotected LLMs - extract system prompts, test instruction conflicts, and learn why simple override attempts sometimes work

25 minutes2 labs

LLM + External Data

Indirect prompt injection through RAG poisoning - how attackers embed malicious instructions in knowledge bases to manipulate LLM outputs

30 minutes1 lab

LLM + Tools

Tool exploitation and excessive agency in LLM systems - discover hidden tools, abuse function calling, and inject through AI-generated output

30 minutes2 labs

LLM + Defenses

Bypass LLM security defenses - keyword filters, instruction hierarchy, self-check prompts, and code-level guards. Learn what works and what doesn't

35 minutes2 labs

LLM Agents & Agentic Security

When AI acts on its own - MCP, A2A, and why agents amplify every vulnerability you've learned

8 minutes3 labs