Excessive agent permissions
A section of the "Security" topic. The questions check an understanding of practice, not memory of specific library APIs. After each answer comes a review: why the correct option is right, what is wrong with each incorrect one and which chapter to read.
Quiz difficulty Formats: several correct options, one correct option, sorting into groups
Quiz 7.6Excessive agent permissionsStart practice Practice has no timer and a review after every answer.
01
What we check
- Three kinds of excessive agency
- A medical database and the model's permissions
- Automatic portfolio rebalancing
- Insecure output handling
- A feature that breaks the law
- Parsing a coding agent's commands
- A support bot's permissions
- A model's answer on a web page
02
Where to read
- Li, AI Agents in Depthsec. 4.6 · sec. 5.1.9
- Wilson, The Developer's Playbook for LLM Securitych. 7 · ch. 7, Securing Your Output Handling · ch. 7, Watch for Excessive Agency