video-generate
Pass
Audited by Gen Agent Trust Hub on Mar 4, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes arbitrary text and image inputs that are passed to external AI models, creating a surface for indirect prompt injection attacks where malicious content could influence the agent's behavior.
- Ingestion points: User inputs provided via the
--promptand--inflags in theagent-mediacommand described inSKILL.md. - Boundary markers: No delimiters or isolation techniques are specified to distinguish untrusted user input from the system's execution context.
- Capability inventory: The skill invokes the
agent-mediaCLI tool, which utilizes network access to interact with well-known video generation services (Fal and Replicate). - Sanitization: There is no evidence of input validation, filtering, or sanitization for the prompts or image sources before they are processed.
Audit Metadata