skills/comfy-org/comfy-skills/comfy/Gen Agent Trust Hub

comfy

Pass

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: All external references and connection endpoints (such as cloud.comfy.org) belong to the official infrastructure of the skill author (comfy-org). These are legitimate vendor resources used for the skill's primary purpose.
  • [SAFE]: The skill provides instructions for managing sensitive API keys using secure methods, such as environment variables and OAuth flows. It uses standard placeholders like comfyui-... for documentation purposes and does not contain any hardcoded credentials.
  • [SAFE]: The shell commands included in the documentation (e.g., openclaw mcp set) are standard configuration steps for setting up a Model Context Protocol server. They are transparent and intended for local environment setup by the user.
  • [SAFE]: The skill describes a remote execution model where image and video generation workflows are processed on the hosted Comfy Cloud service rather than the local machine, which minimizes the local security footprint.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 22, 2026, 11:53 PM