mine-competitor-reviews
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface by processing untrusted app store reviews that may contain malicious instructions.
- Ingestion points: Publicly available review text retrieved from app stores as described in SKILL.md.
- Boundary markers: The skill contains explicit instructions to treat review text as untrusted data and specifies that embedded links or instructions must never authorize execution, tool calls, or disclosure.
- Capability inventory: The agent has the capability to write research reports to a local research folder.
- Sanitization: The instructions mandate the exclusion of reviewer names and identifiers, and the redaction of personal information encountered during processing.
Audit Metadata