advibly-vox-explainer
Pass
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for a local workflow using
ffmpeg,curl, andffprobeto mix generated media assets. This includes a complex filter for sidechain-ducked audio and normalization. - [EXTERNAL_DOWNLOADS]: The workflow includes
curlcommands to download media components generated by the platform's tools for final local assembly. - [PROMPT_INJECTION]: The skill ingests user-provided text for brand angles and product information, which is then used to generate prompts for various media tools.
- Ingestion points: User input for brand claims, product details, and narration scripts.
- Boundary markers: Uses structural prompt blocks and quotes for headlines, though no explicit sanitization is performed on user strings.
- Capability inventory: Accesses tools for image, video, voice, and music generation, and generates local shell commands.
- Sanitization: Lacks specific filtering or escaping mechanisms for external content before interpolation.
Audit Metadata