ai-video-generation
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
beltCLI tool, executed via Bash as restricted by the skill configuration, to perform all video generation and management tasks. - [INDIRECT_PROMPT_INJECTION]: The skill acts as an interface for processing external data and user-provided prompts, creating an attack surface for instructions embedded in ingested media or text.
- Ingestion points: Prompts and remote media URLs (e.g.,
image_url,audio_url,video_url) provided as input to the CLI. - Boundary markers: Inputs are structured within JSON objects, providing a layer of separation between data and command arguments.
- Capability inventory: The skill has the ability to execute shell commands via the
beltCLI as defined in the allowed-tools configuration. - Sanitization: The instructions do not specify validation or sanitization procedures for the external content or URLs before they are processed by the tool.
- [EXTERNAL_DOWNLOADS]: The skill references external resources for installation and documentation purposes.
- Fetches installation guidelines from the
inference-shGitHub repository. - Links to application assets and documentation hosted on the
inference.shdomain.
Audit Metadata