scenario-ad-variants
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate instructions for video editing and campaign management using a suite of dedicated tools.
- [COMMAND_EXECUTION]: The mention of an
npxcommand is an instruction for the user to manually install sibling skills and does not involve autonomous shell execution by the agent. - [DATA_EXFILTRATION]: Asset handling and downloading are constrained to the skill's functional requirements for media processing, with no evidence of sensitive data harvesting or unauthorized network transmission.
- [INDIRECT_PROMPT_INJECTION]: The skill includes an attack surface for indirect injection as it processes user-provided briefs and assets to construct generative prompts. However, this is inherent to its primary purpose as a video editing assistant, and the skill provides guidelines for inspecting outputs to ensure they align with the source.
Audit Metadata