ai-video-gen
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill contains parameters for user-provided prompts and image URLs which are sent to external APIs. While this constitutes an injection surface, it is fundamental to the skill's purpose and handled through standard API integration patterns.
- Ingestion points: prompt and reference_image_url in SKILL.md.
- Boundary markers: None specified in the documentation.
- Capability inventory: Network requests (POST/GET) to api.heygen.com and other video provider endpoints.
- Sanitization: Not explicitly defined within the integration instructions.
- [EXTERNAL_DOWNLOADS]: The skill facilitates network requests to official and well-known service endpoints including api.heygen.com and fal.ai for video generation tasks.
- [COMMAND_EXECUTION]: The provided curl commands are standard examples for API testing and do not demonstrate malicious intent or unauthorized access.
Audit Metadata