Skills
skills/agentspace-so/runcomfy-skills/flux-2-klein/Gen Agent Trust Hub

flux-2-klein

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute the runcomfy CLI tool for image generation and authentication (runcomfy login, runcomfy run).
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the @runcomfy/cli Node.js package and downloads generated image assets from *.runcomfy.net and *.runcomfy.com domains.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted user data (text prompts) which are then interpolated into a JSON payload for a CLI command.
  • Ingestion points: User prompts are captured in the prompt field within SKILL.md examples.
  • Boundary markers: The documentation explicitly states that the CLI does not shell-expand the prompt and transmits the JSON body directly to prevent injection.
  • Capability inventory: The skill uses subprocess calls to the runcomfy CLI and performs network operations to the RunComfy API.
  • Sanitization: The skill documentation claims the CLI handles prompt transmission securely, though the skill itself does not implement additional sanitization logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 09:25 AM