AI Agents
Agent Security and Guardrails
Agents can take actions, so mistakes matter more. Permissions, sandboxing, approval steps and injection defence.
- Advanced
- About 6 hours
- 1 modules
- 6 lessons
- Free
This course is not released yet. We publish courses one at a time, each one complete. The full syllabus is below.
What you will learn
- SafetyRisks of Autonomous Actions, Least-Privilege Tools, Sandboxing and more
Course syllabus
Not released yet · full syllabus below
- 1
Safety
6 lessons- 1Risks of Autonomous ActionsSoon
- 2Least-Privilege ToolsSoon
- 3SandboxingSoon
- 4Approval GatesSoon
- 5Prompt Injection in AgentsSoon
- 6Input and Output GuardrailsSoon