cli-anything-audacity
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external audio files and project JSON data, which could serve as a vector for indirect prompt injection attacks.
- Ingestion points: Commands such as
project open,clip import, andmedia probeingest metadata and content from external files into the agent's context as specified inSKILL.md. - Boundary markers: The instructions lack specific boundary markers or warnings to the agent to disregard instructions potentially embedded in the processed media.
- Capability inventory: The CLI possesses the capability to modify the local file system through commands like
project saveandexport render. - Sanitization: There is no documentation of metadata sanitization or validation for the audio files or project JSON handled by the skill.
Audit Metadata