shot-transitions
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: Workflow files such as
workflows/seam_ltx25_flf2v_transition.api.jsoningest images viaLoadImagenodes and prompts viaPrimitiveStringMultilineorCLIPTextEncodenodes. Thescripts/comfy_client.pyscript also allows overriding workflow inputs via the--setcommand-line argument. - Boundary markers: No explicit delimiters or instructions to ignore embedded content within processed data were identified.
- Capability inventory: The skill possesses network access capabilities through
urllib.requestinscripts/comfy_client.pyand file system write capabilities to save generated outputs. - Sanitization: Input validation is limited to basic JSON parsing of overrides; no advanced sanitization of image metadata or prompt text is performed before transmission to the ComfyUI backend.
- [COMMAND_EXECUTION]: The skill includes Python scripts (
scripts/comfy_client.py,scripts/preflight.py) designed to be executed by the user to interact with an external ComfyUI API. - The scripts facilitate network communication and file management (uploading inputs, downloading generated videos).
- Sensitive data such as the
COMFY_API_KEYis managed via environment variables, following standard security practices.
Audit Metadata