Learning Platform

Master AI Security

Free, in-depth guides that take you from the fundamentals to advanced red teaming and governance — with theory, example architectures, and code you can adapt.

Learning Path

Read them in order for a full curriculum, or jump to the topic you need. Every guide is free — no signup required.

Beginner18 min read

AI Security Fundamentals

A ground-up introduction to securing AI and LLM systems: the trust boundaries that matter, the attack surface of a modern GenAI app, and a reference architecture you can defend.

Read the guide

What you'll learn

Why AI security is different
The GenAI trust boundary
Attack surface of an LLM app
A defensible reference architecture
Your first threat model

Intermediate24 min read

LLM Security & Prompt Defense

A deep, practical guide to prompt injection, jailbreaking, and building layered guardrails — with a production-ready defense architecture and code you can adapt.

Read the guide

What you'll learn

Direct vs. indirect injection
How jailbreaks actually work
The guardrail stack
A production defense architecture
Testing your defenses

Intermediate20 min read

Securing RAG Pipelines

Retrieval-augmented generation turns your knowledge base into part of the prompt — and part of the attack surface. A practical guide to the RAG threat model, indirect injection, and a defensible ingestion-to-answer architecture.

Read the guide

What you'll learn

Why RAG expands the attack surface
The RAG threat model
Indirect injection through retrieved content
A defensible RAG architecture
Access control and tenant isolation
Testing a RAG pipeline

Advanced22 min read

GenAI Red Teaming

An offensive-security methodology for GenAI systems: how to plan, execute, and report an LLM red-team engagement — plus a repeatable harness architecture for automated testing.

Read the guide

What you'll learn

What GenAI red teaming is
A repeatable methodology
Attack techniques by objective
An automated red-team harness
Scoring and reporting

Specialist24 min read

AI Agent & MCP Security

When an LLM can call tools, a prompt injection stops being a bad answer and becomes a real action — deleted data, spent money, exfiltrated secrets. A practical guide to securing tool-calling agents and the Model Context Protocol (MCP).

Read the guide

What you'll learn

From chatbots to agents
Excessive agency and the confused deputy
The tool-definition threat model
MCP-specific risks
A defensible agent architecture
Auditing your tools

Specialist20 min read

AI Compliance & Governance

How to turn AI security into a defensible program: the major frameworks (EU AI Act, NIST AI RMF, ISO/IEC 42001), a governance architecture, and a control mapping you can operationalize.

Read the guide

What you'll learn

Security vs. governance
The frameworks that matter
A governance operating model
Mapping controls to technical checks
Evidence and audit readiness

More free resources

Available right now — no signup required.

Put it into practice

Guides are only half the work. Run your own prompts and agent tools through the free Analyze and Agent Guard tools, then try the attacks yourself in the Red-Team Playground.

Open Analyze Red-Team Playground

Master AI Security

Learning Path

AI Security Fundamentals

What you'll learn

LLM Security & Prompt Defense

What you'll learn

Securing RAG Pipelines

What you'll learn

GenAI Red Teaming

What you'll learn

AI Agent & MCP Security

What you'll learn

AI Compliance & Governance

What you'll learn

More free resources

OWASP Top 10 for LLMs

Red-Team Playground

Blog

Docs

Put it into practice