pixar-ad
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the execution of several shell commands, including 'ffmpeg' and 'ffprobe' for trimming and concatenating video beats, as well as 'curl' for API calls. These commands are necessary for the primary purpose of video ad assembly and deployment.
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its handling of external product data. (1) Ingestion point: Gate 1 in 'SKILL.md' accepts product URLs and listing text. (2) Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the ingested data. (3) Capability inventory: The skill can perform network operations via 'curl' and manipulate local files via 'ffmpeg' and 'python3'. (4) Sanitization: No validation or filtering is applied to the ingested external content before it is interpolated into prompts.
- [DYNAMIC_EXECUTION]: The skill invokes a shared Python utility ('shared/skills/music-mix/scripts/music_mix.py') to perform audio mastering and level verification at Gate 6.
- [EXTERNAL_DOWNLOADS]: The skill uses 'curl' to communicate with 'api.novoads.ai' and to download media assets. These operations target the vendor's own infrastructure and represent intended functionality.
Audit Metadata