ai-video-gen
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill accepts untrusted user input via the
--promptargument and image URLs, which are subsequently transmitted to external AI video generation models. - Ingestion points: User-supplied prompts, image paths or URLs, and visual style preferences retrieved from
style.mdprofile data. - Boundary markers: The provided instructions do not include specific delimiters or 'ignore embedded instructions' warnings to isolate user data from system prompts.
- Capability inventory: The skill executes a local Python script (
ai_video.py) that performs network operations (API calls to providers) and writes generated files to the local file system (using the-ooutput flag). - Sanitization: There is no indication of input validation, escaping, or prompt sanitization before the content is passed to external providers.
- [COMMAND_EXECUTION]: The skill's core functionality is implemented through the execution of a local script,
skills/shared/scripts/ai_video.py, via the Python interpreter. This script is used for configuration verification, capability probing, and managing the asynchronous video generation lifecycle (submission, polling, and downloading).
Audit Metadata