product-context
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external web content and user-provided data. It implements comprehensive safety instructions to treat all retrieved content strictly as data, never as instructions. It specifically directs the agent to report attempted injections as findings rather than obeying them, providing a strong defense against adversarial web content.
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill is intended to collect and store business context. It includes strict mandates to identify and flag credentials for rotation without reproducing them, and explicitly forbids the persistence of special-category personal data. It utilizes research tools like WebFetch within restricted, user-defined boundaries.
- [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The documentation includes a helper shell command (grep) for the user to audit their own filesystem. This is presented as an informational tip for manual use and is not executed by the agent, nor are shell tools requested in the skill tool configuration.
Audit Metadata