ai-video-generation
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a legitimate wrapper for the Pexo video generation platform, providing tools to automate the workflow between the agent and the Pexo backend.
- [COMMAND_EXECUTION]: The skill uses shell scripts to perform API requests via
curland process JSON data withjq. All commands are scoped to the skill's intended purpose and target the official vendor domain (pexo.ai). - [DATA_EXFILTRATION]: User-provided prompts and media files are uploaded to Pexo's infrastructure for processing. This behavior is essential for the skill's functionality and is directed exclusively to the vendor's verified endpoints.
- [PROMPT_INJECTION]: Analysis of the instructions in
SKILL.mdandREADME.mdshows no attempts to bypass safety filters or override system prompts. Guidance to relay user input 'verbatim' is a functional requirement for the remote video agent to receive accurate creative instructions.
Audit Metadata