flux-image

Pass

Audited by Gen Agent Trust Hub on Apr 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issue s were identified. The skill facilitate s interaction with the official inference.sh platform using authorize d tools and documentation.- [COMMAND_EXECUTION]: The skill use s the infsh CLI tool. This is properly constraine d in the skill configuration to prevent arbitrary shell com m and execution.- [EXTERNAL_DOW NLOADS]: The skill reference s setup documentation and modular com ponents hosted on the vendor's official GitHub repository (inference-sh/skills).- [PR OMPT_INJECTION]: The skill accept s use r inpu t for prom pts.
  • Ingestion points: Use r-provide d text prom pts and image URLs in SKILL.md.
  • Boundary marker s: None explicitly define d.
  • Capability inventory: Bash(infsh *) as define d in the skill frontmatter.
  • Sanitization: None; inpu t is passed directly to the CLI com m and. The surface is standard for image generation functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 17, 2026, 09:28 AM
Security Audit — agent-trust-hub — flux-image