> 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 keeps conversations, files, and other private data tied to the right user.

# Private data in AI agents


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

An agent may remember a conversation, retrieve a file, or look up a personal record on a user's behalf. This scenario checks that those features respect the current user's permissions when they handle data belonging to someone else.

Use it for agents that store or retrieve user-specific information. A finding points to the type of data the agent disclosed or made available outside its owner's account.

The app lists this scenario as **Read another user's private agent data** (`AG-01`) in [Agent Red Teaming](https://www.superagent.sh/docs/products/red-teaming/agents).

## Standards and guidance

[OWASP LLM02:2026 Sensitive Information Disclosure](https://genai.owasp.org/resource/owasp-genai-llm-top-10-2026/)

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