pets-and-animals-lead
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [SAFE]: The skill implements a legitimate workflow for video prompt engineering without any detected malicious intent or dangerous command execution.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from user inputs and reference cases in 'references/cases.md'. Evidence chain: (1) Ingestion points: User input and external prompt excerpts in references/cases.md. (2) Boundary markers: Present via a required 'Preserve/Replace/Avoid' reference contract and a final verification checklist in SKILL.md. (3) Capability inventory: Text generation and media inspection tools; no file-writing or subprocess capabilities are utilized by the scripts. (4) Sanitization: Explicit instructions to distinguish evidence from recommendations and to avoid superficial copying.
- [EXTERNAL_DOWNLOADS]: The workflow involves fetching media from 'goodcase.ai' and 'x.com' for visual reference. These operations are consistent with the skill's primary purpose and do not involve executable content or sensitive data access.
Audit Metadata