qwen-image-2
Pass
Audited by Gen Agent Trust Hub on Mar 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill interacts with the AI model through the
infshCLI. The agent's capabilities are restricted to this specific tool via theallowed-toolspolicy, adhering to the principle of least privilege. - [EXTERNAL_DOWNLOADS]: Setup instructions refer to fetching the
inference-sh/skillspackage usingnpx. This originates from a well-known service provider associated with the skill's purpose. - [PROMPT_INJECTION]: The skill ingests user text for image generation, representing an indirect injection surface. Ingestion points:
promptandnegative_promptfields in JSON inputs. Boundary markers: Data is structured in JSON format. Capability inventory: Execution ofinfshin a shell. Sanitization: Handled by the backend model providers.
Audit Metadata