# Using PyRIT for Adversarial Testing

> Master Microsoft's PyRIT framework to orchestrate complex, multi-step adversarial attacks and evaluate LLM resilience against sophisticated jailbreaks.

Canonical: https://codelivly.com/rooms/using-pyrit-for-adversarial-testing
Last updated: 2026-10-14

| | |
|---|---|
| Level | Advanced |
| Category | Cybersecurity |
| Lessons | 3 |
| Estimated time | 30 min |
| XP | 100 |

This module provides a comprehensive guide to PyRIT (Python Risk Identification Toolkit for AI), an open-source framework designed for orchestrating complex, multi-step adversarial attacks against AI systems. You will learn how to configure PyRIT's objective functions, leverage its diverse attack techniques (like Crescendo and PAIR), and automate the evaluation of LLM resilience against sophisticated jailbreaks and multi-turn prompt injections.

## Lessons

1. What is PyRIT and How It Automates Attacks (30 min)
2. Configuring PyRIT Against a Target (30 min)
3. Building Custom Attack Strategies in PyRIT (30 min)

## Related

- [Learning path: Automated AI Security Testing](https://codelivly.com/learning-paths/automated-ai-security-testing)
- [All rooms](https://codelivly.com/rooms)