youtube-video-generation
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external, untrusted content (such as video scripts and user prompts), which creates a potential surface for indirect prompt injection attacks.\n
- Ingestion points: Data enters the agent's context through the
messagesfield in thecurlexamples documented inSKILL.md.\n - Boundary markers: The skill lacks explicit boundary markers or specific instructions to the underlying model to ignore or separate potentially malicious commands embedded within user-provided scripts.\n
- Capability inventory: The skill utilizes network capabilities to communicate with the vendor's API at
eachsense-agent.core.eachlabs.run.\n - Sanitization: There is no documentation regarding the sanitization, validation, or filtering of the inputs before they are processed by the media generation engine.
Audit Metadata