AI Agents, Integrations and Human Escalation
Design agents with bounded tools, explicit permissions and reliable handoff when they cannot safely proceed.
What you will build
An agent contract, permission map and adversarial test plan.
Before you start: Advanced Automation and Exception Handling, AI Output Evaluation and Quality Assurance.
74 min estimated, including practice. 4 written lessons.
What you will learn
- Bound the agent’s authority
- Treat external content as untrusted input
- Design approvals and escalation payloads
- Evaluate the agent as a system
Course lessons
- Bound the agent’s authority
- Treat external content as untrusted input
- Design approvals and escalation payloads
- Evaluate the agent as a system
Complete the lessons and pass the five-question knowledge assessment at 80% to earn a free course certificate. Exercises are self-directed, not independently graded.
Download the practice workbookSources and further reading
Original Academy teaching and fictional examples. These references provide context, not endorsement. Edition 2026.09; updated 2026-09-24.
- OWASP: Top 10 for large language model applications
Security background for applications that use language models.
- NIST: AI Risk Management Framework
Voluntary framework for organising AI risks and controls.
- GOV.UK: Responsible AI in recruitment
UK guidance on procuring and deploying recruitment AI.