Skills
Navigate
skills/rafaelgrasa/skills/qwen-image-2/Gen Agent Trust Hub

qwen-image-2

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides links to external installation documentation hosted on GitHub (github.com/inference-sh) for the required infsh command-line tool.
  • [COMMAND_EXECUTION]: Instructs the agent to perform operations such as infsh login and infsh app run using the Bash tool to interact with the image generation models.
  • [DATA_EXFILTRATION]: Transmits user-provided prompts and image metadata to the inference.sh service to perform model inference, which is the intended purpose of the skill.
  • [PROMPT_INJECTION]: Ingests untrusted user input via the prompt and reference_images fields which are then processed by the agent.
  • Ingestion points: SKILL.md input parameters
  • Boundary markers: Data is encapsulated within a structured JSON object for the CLI command
  • Capability inventory: Shell command execution via Bash(infsh *) in SKILL.md
  • Sanitization: Relies on the agent's ability to correctly escape JSON strings when invoking the Bash tool
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 11:26 PM