comfyui-troubleshooter

Pass

Audited by Gen Agent Trust Hub on Feb 24, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [NO_CODE]: The skill consists entirely of a Markdown file (SKILL.md) containing documentation and diagnostic logic. It does not include any Python, JavaScript, or shell scripts.
  • [EXTERNAL_DOWNLOADS]: The documentation includes instructions for users to download models from Hugging Face and install custom nodes from GitHub. These refer to well-known technology services and are presented as neutral troubleshooting steps for the user to perform manually.
  • [PROMPT_INJECTION]: The skill defines a process for the agent to ingest external data such as error logs and workflow descriptions (Category 8 surface). Since the skill is informational and lacks the capability to automatically execute code or exfiltrate data, the risk associated with this surface is negligible.
  • Ingestion points: Step 2 describes gathering error messages and workflow content from the user or system logs.
  • Boundary markers: Not defined in the current troubleshooting logic.
  • Capability inventory: The skill suggests manual commands (git clone, huggingface-cli) and provides guidance on interacting with local ComfyUI API endpoints (/interrupt, /free).
  • Sanitization: None identified as the skill is documentation-based.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 24, 2026, 08:38 AM