drive-app
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill directs the agent to read, follow, and update instructions from a repository file (HOW-TO-USE-THE-PRODUCT.md). This file serves as an external source of instructions that could be manipulated to perform unauthorized actions.
- Ingestion points: The agent is instructed to read the guide HOW-TO-USE-THE-PRODUCT.md at the root of the repository.
- Boundary markers: The instructions do not provide boundary markers or delimiters to help the agent distinguish between the skill's trusted instructions and the potentially untrusted content of the guide.
- Capability inventory: The agent has the capability to start local services, interact with sign-in flows, manage test accounts, and use test payment modes.
- Sanitization: There are no requirements to sanitize or validate the content of the guide before the agent acts upon it.
Audit Metadata