whiteboard-video
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
whiteboard-video-enginePython package from the author's official GitHub repository (gnipbao/whiteboard-video-engine). - [COMMAND_EXECUTION]: The skill's primary functionality is delivered through a Python CLI wrapper (
whiteboard_cli.py) that executes rendering and processing commands. - [DYNAMIC_EXECUTION]: The skill is designed to auto-discover and execute model wrapper scripts from the
tools/lineart/directory relative to the current working directory, allowing for project-specific model integration. - [INDIRECT_PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by processing external markdown scripts and scene plans to generate video content. (1) Ingestion points: Commands such as
runandplan-scriptingest external markdown files (e.g.,examples/ten-second-demo.md) to define video scenes. (2) Boundary markers: The instructions do not specify delimiters or warnings to ignore instructions within the processed data. (3) Capability inventory: The skill calls a CLI that performs file system operations, video rendering, and potential network requests for TTS generation. (4) Sanitization: No input sanitization or validation of the ingested script content is explicitly mentioned.
Audit Metadata