comfyui
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using
curlandbase64to interact with the ComfyUI API and process image files for transmission. - [EXTERNAL_DOWNLOADS]: The skill performs network requests to download generated image files from the user-configured
COMFYUI_API_SERVERto the local filesystem. - [DATA_EXFILTRATION]: The skill transmits data externally, including authentication tokens and Base64-encoded image content. While this is the intended functional behavior, it establishes a network communication surface for data transfer.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it interpolates untrusted user-provided descriptions into API request bodies and image generation prompts.
- Ingestion points: User-provided image descriptions in
SKILL.mdand associated references. - Boundary markers: None identified; user input is processed without delimiters.
- Capability inventory: Network access and file write capabilities via
curl, and file read capability viabase64. - Sanitization: No validation or escaping of external content is specified before interpolation into prompts.
Audit Metadata