video-replace-audio-api
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill manages its API key by storing it in a dedicated file (~/.pixazo/api-key) with restrictive permissions (chmod 600), which is a recommended practice for secret management.\n- [SAFE]: All network operations, including API calls and documentation references, are directed to domains owned by the skill's vendor (pixazo.ai).\n- [INDIRECT_PROMPT_INJECTION]: The skill acts as a surface for processing user-provided data (prompts and media URLs) sent to a remote API. This is an inherent design characteristic for media processing tools.\n
- Ingestion points: User-provided video/audio URLs and text prompts processed in SKILL.md.\n
- Boundary markers: None explicitly defined in the API request construction.\n
- Capability inventory: Network POST requests to gateway.pixazo.ai.\n
- Sanitization: None specified in the instructions; assumes backend API processing.
Audit Metadata