01
advanced60 min
OWASP LLM Top 10 ; The Attack Surface Map
Every item in the OWASP LLM Top 10 with real exploit techniques, severity context, and what a red teamer actually does with each one.
owaspllm01prompt-injectionattack-surfacered-team
All modulesModule 04
Offensive AI
The OWASP LLM Top 10, prompt injection weaponization, jailbreaking taxonomies, agent hijacking, and the MCP attack frontier.
5 lessons330 min total
02
advanced70 min
Prompt Injection ; Techniques and Weaponization
The primary AI attack class: direct injection, indirect injection, multi-modal injection, and building injection test suites that cover real attack scenarios.
prompt-injectionindirect-injectionattackweaponizationtesting
03
advanced65 min
Jailbreaking ; Methods and Red Team Automation
Systematic jailbreaking: attack categories, automated red teaming with garak and PyRIT, and what actually works on production models.
jailbreakgarakpyritautomated-red-teamsafety-bypass
04
advanced70 min
Agent Hijacking ; Attacking Autonomous Systems
When agents can take actions, prompt injection becomes code execution. Goal hijacking, memory poisoning, tool abuse, and the cascading failure problem in multi-agent systems.
agent-hijackinggoal-hijackingmemory-poisoningtool-abuseautonomous
05
advanced65 min
MCP and Tool Security ; Attacking AI Integrations
The emerging attack frontier: malicious MCP servers, tool schema poisoning, rug-pull attacks, and why connecting your AI to the internet is harder than it looks.
mcptool-securitysupply-chainrug-pullintegration