faceswap
Pass
Audited by Gen Agent Trust Hub on Nov 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill accepts external URLs as input for face swapping, which creates a surface for indirect prompt injection if the remote content is used to influence agent behavior.
- Ingestion points:
source_image_urlandtarget_video_urlparameters in the workflow execution requests inSKILL.md. - Boundary markers: None provided in the implementation examples.
- Capability inventory: Network POST and GET operations to a remote API for video processing.
- Sanitization: No sanitization or validation logic is present in the provided code snippets.
- [EXTERNAL_DOWNLOADS]: The skill interacts with the HeyGen API (api.heygen.com), a well-known and established service for AI video generation.
Audit Metadata