blender-video-original
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions direct the agent to fetch additional dependencies from the author's own repository (full-aigc-skills/blender-skills) using the npx package runner.
- [REMOTE_CODE_EXECUTION]: The use of 'npx skills add' to install and execute tools from a remote repository at runtime constitutes remote code execution.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-supplied video concepts to generate structured plans and executable Blender scenes, creating an attack surface for indirect prompt injection.
- Ingestion points: User-supplied prose video concepts described in SKILL.md.
- Boundary markers: None; the instructions do not specify delimiters or safety warnings for the agent when processing the input concept.
- Capability inventory: File system writes (shot-plan.md, .blend files), scene mutation within Blender, and tool installation via npx.
- Sanitization: No validation or sanitization of the input text is required before it is interpolated into the workflow.
Audit Metadata