aliyun-qwen-image-edit-test
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill includes an executable bash example that invokes a local Python script (
prepare_edit_request.py) located within the skill's directory structure. This is used for generating request configurations. - [EXTERNAL_DOWNLOADS]: The provided example uses
example.comfor an input image URL, which is a standard, safe placeholder domain. - [PROMPT_INJECTION]: The skill provides an interface for user-defined prompts via a command-line argument. Ingestion points:
SKILL.md. Boundary markers: None. Capability inventory: Local script execution. Sanitization: None. This is categorized as an attack surface for indirect prompt injection, though standard for this use case.
Audit Metadata