video-generation
Pass
Audited by Gen Agent Trust Hub on May 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill acts as a controller for video generation workflows, providing classification rules and routing tables to hand off tasks to other specialized skills or runners.
- [COMMAND_EXECUTION]: The skill references the
postplusCLI tool for diagnostics (postplus doctor) and quota management (postplus quote confirm). These commands are associated with the vendor's own infrastructure (PostPlusAI) and are used for legitimate operational tasks within the ecosystem. - [DATA_EXPOSURE]: The usage of
--challenge-file <challenge.json>in the command boundary is a mechanism for service-level authentication/confirmation and does not involve accessing sensitive user credentials or system files. - [PROMPT_INJECTION]: The skill implements architectural constraints through 'Core Boundary' and 'Three-Layer Boundary' definitions, which explicitly limit the agent's scope to classification and routing, preventing unauthorized execution or creative inference.
Audit Metadata