comfyui-video-pipeline
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill body consists of legitimate technical documentation for orchestrating video generation workflows. No signs of prompt injection, persistence, or privilege escalation were detected.- [SAFE]: Frontmatter configuration (e.g., COMFYUI_URL, curl, wget) is appropriate for the skill's stated purpose of interacting with a ComfyUI environment.- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data in the form of text descriptions and character images to produce video content.
- Ingestion points: User-supplied text and images processed via the pipelines described in SKILL.md.
- Boundary markers: None explicitly mentioned in this documentation.
- Capability inventory: The skill frontmatter requires access to binary tools (curl, wget) and interacts with an external API specified by the COMFYUI_URL environment variable.
- Sanitization: Not specified within the instructional content. (Note: This finding documents the inherent attack surface of a generative AI skill and does not indicate a specific vulnerability or malicious intent.)
Audit Metadata