seedance-short-drama
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands using
bashto interact with a local script../ofox-video-core/references/ofox-video.sh. This is a standard delegation pattern where one skill relies on a core utility skill from the same vendor. - [EXTERNAL_DOWNLOADS]: The skill downloads generated video files from the Ofox API (app.ofox.ai). It also provides instructions for installing the skill suite via
npx skills add ofoxai/skills, which is a recognized method for managing agent extensions. - [INDIRECT_PROMPT_INJECTION]: As a generative tool, this skill processes untrusted user data (scripts and dialogue) to construct prompts for an external AI video model. While it provides guidance on prompt structure, it remains a surface for indirect injection attacks targeting the downstream video model.
Audit Metadata