krea

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a documentation repository for integrating Krea (hosted API and open weights) into ComfyUI workflows. It contains architectural guidance, pricing matrices, and model weight measurements verified against primary sources.
  • [EXTERNAL_DOWNLOADS]: The documentation references several external resources, including official Hugging Face repositories for Black Forest Labs and Comfy-Org, and community-developed custom-node packs on GitHub. The author provides clear differentiation between established tools and unproven ones (e.g., highlighting that 'eliteprox/ComfyUI-Krea' has 0 stars and is unproven), encouraging safe usage practices.
  • [PROMPT_INJECTION]: Instructions focus on technical selection logic (e.g., choosing Hosted API vs. Local Weights based on job requirements) and do not contain attempts to override agent safety filters or system instructions.
  • [DATA_EXFILTRATION]: No evidence of hardcoded credentials, sensitive file access, or unauthorized network operations was found. The mentioned API paths (e.g., /proxy/krea/generate) are standard ComfyUI proxy routes for partner nodes.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 09:01 AM
Security Audit — agent-trust-hub — krea