ads-setup
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data such as business models, offers, and website content to generate profiles. It mitigates this by explicitly instructing the agent to treat websites and uploaded material as untrusted data.
- [DATA_EXPOSURE]: The skill provides explicit guidelines for secret management, directing the agent to store non-secret references only and to use environment variables or OS keychains instead of storing actual API keys, tokens, or passwords in generated artifacts.
- [REMOTE_CODE_EXECUTION]: The skill contains a specific security boundary that refuses remote pipe-to-shell installation methods (e.g.,
curl | bash). It mandates the use of authenticated local checkouts and SHA-256 checksum verification for any installations.
Audit Metadata