flux-image

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides links to official installation instructions and guidelines hosted on the inference.sh GitHub repository.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external data, specifically user-defined prompts and image URLs, for AI image generation, creating a surface for indirect instructions.
  • Ingestion points: User-provided text prompts and image URLs within the JSON input for the belt CLI commands in SKILL.md.
  • Boundary markers: The skill utilizes structured JSON to delimit input data from the command execution.
  • Capability inventory: The skill executes shell commands via the belt CLI tool using the Bash environment.
  • Sanitization: No explicit sanitization, content filtering, or validation logic for the input prompts is defined within the skill file.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 05:54 PM
Security Audit — agent-trust-hub — flux-image