> For clean Markdown of this page, append .md to its URL. For the complete documentation index, see https://www.superagent.sh/llms.txt.


Check whether an AI agent enforces role restrictions on sensitive actions and tools.

# Privileged actions in AI agents


[← Back to attack scenarios](https://www.superagent.sh/docs/products/red-teaming/attack-scenarios)

Some agent tools are available only to administrators or other privileged users. This scenario checks whether those limits still hold when a user asks the agent to act for them.

Choose it when people with different roles use the same agent. The resulting evidence can help you locate the role or permission check that failed.

The app lists this scenario as **Escalate from ordinary user to privileged agent actions** (`AG-03`) in [Agent Red Teaming](https://www.superagent.sh/docs/products/red-teaming/agents).

## Standards and guidance

[OWASP Agentic Top 10 (2026), ASI03 Identity & Privilege Abuse](https://genai.owasp.org/resource/owasp-top-10-for-agentic-applications-for-2026/)

---
Source: https://www.superagent.sh/docs/products/red-teaming/attack-scenarios/agent-privilege-escalation
Index: https://www.superagent.sh/llms.txt
