Excessive Agency
Testing the gap between what an agent is capable of doing and what it actually needs to do to complete its intended task.
- Category
- Agentic AI Security
- Difficulty
- Intermediate
- Last Updated
- 20 September 2026
Lab detail
Objective
“Excessive agency” is granting an agent broader tool access, autonomy or permission than its actual task requires — a design failure that widens every other agentic risk in this catalog. This lab will take a reference agent built for a narrow task and measure how much unrelated action becomes possible purely because of over-broad tool grants, independent of any specific manipulation technique.
Planned Scope
- Define a narrow reference task and the minimal tool set it actually requires.
- Build a version of the agent with intentionally broader tool access, matching common real-world over-provisioning.
- Enumerate the unrelated actions that broader access makes possible.
- Document a least-privilege tool-scoping methodology for agent design.
Framework Mapping
Referenced frameworks
Exact control-ID mappings are to be validated as this lab progresses — see the full Frameworks reference.
Responsible Use / Safety Notes
How this lab is run
This lab uses synthetic data and controlled environments. It is intended to demonstrate defensive AI-security principles and does not contain customer data, production credentials or intentionally weaponized exploit infrastructure.
A decision is waiting for sign-off. Get an independent view first.
Tell me what is being decided and what is at stake. I will first say honestly whether my experience fits — then we talk about scope, not before.
Or connect on LinkedIn first.