Bound the agent’s authority
Course overview · 4 min reading + 12 min practice, estimated
Principles and method
An agent combines model reasoning with tools and state to pursue a task. The key design question is not what it can say but what it can do. Define the goal, allowed tools, data scope and prohibited actions. Separate read access from write access and drafting from external communication. Use narrow permissions and enforce them in the tool layer, not only in a prompt. Set limits on iterations, cost and scope. A recruitment research agent should not automatically inherit authority to contact people, change assessments or disclose records.
Worked example
An agent may retrieve approved company facts and draft a research brief. Sending messages and changing application decisions require separate authorised actions that the agent’s tools do not permit by default.
Put it into practice
Write an agent contract for a fictional research assistant with five allowed and five prohibited actions.
Use fictional information and keep your work in your own notes.
Compare your approach: self-review guidance
Describe concrete tool permissions and record scope. Do not rely on a sentence saying be careful while providing an unrestricted send or database-write tool.
Sources 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.