flux-image

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references installation guidelines and model configurations hosted on the official inference.sh GitHub repository, which is a recognized service provider for AI inference.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided text prompts to generate images, which is an inherent part of its functionality. The risk is minimized by the use of structured JSON for inputs and a restricted execution environment.
  • Ingestion points: Prompt and image_url fields within JSON objects in SKILL.md.
  • Boundary markers: Data is encapsulated within JSON structures.
  • Capability inventory: Command execution is strictly limited to the 'belt' CLI tool via the allowed-tools configuration.
  • Sanitization: Input processing is managed by the backend inference service.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:23 PM
Security Audit — agent-trust-hub — flux-image