magnific-video-gen
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests raw user creative ideas verbatim to generate video prompts, which could allow instructions embedded within user data to influence the agent's behavior.
- Ingestion points: The
promptparameter in thevideo_plancall and theclips[].promptparameter invideo_generatewithinSKILL.md. - Boundary markers: The skill instructions do not specify any delimiters or safety warnings to ignore instructions that might be contained within the user-provided prompt strings.
- Capability inventory: The skill utilizes a suite of Magnific MCP tools for video planning, generation, upscaling, and account management as specified in the
allowed-toolsmetadata. - Sanitization: No input validation, filtering, or escaping is performed on the user-supplied prompts before they are passed to the generation tools.
Audit Metadata