dreamina-canvas-generate-video
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
dreamina-canvasCLI tool to create and edit video nodes based on user parameters, representing the primary functionality of the skill. - [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by incorporating untrusted user input directly into shell command arguments without explicit sanitization or delimiters.
- Ingestion points: User-provided content for the
<prompt>and<transition prompt>placeholders in command templates found inSKILL.mdandreferences/video-node-contract.md. - Boundary markers: Absent; the skill does not instruct the agent to use delimiters or ignore embedded instructions within the user input.
- Capability inventory: Execution of the
dreamina-canvasCLI tool for node creation and modification, as documented inSKILL.md. - Sanitization: No sanitization, validation, or escaping mechanisms are specified for the user-supplied prompt data.
Audit Metadata