alibabacloud-wuying-console
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes Python scripts (
check_environment.pyandgenerate_command_catalog.py) to execute the officialaliyunCLI for version checking, plugin management, and schema discovery. These executions are restricted to the vendor's own official tools and are necessary for the skill's primary management functions. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface where it processes command descriptions and help text from local Aliyun CLI plugins to build a command catalog.
- Ingestion points:
scripts/generate_command_catalog.pyreads output fromaliyun <product> --help-all, andSKILL.mdinstructs the agent to inspect live help strings. - Boundary markers: The skill uses structured JSON cataloging and instructs the agent to match intent against official command names and descriptions rather than executing raw text.
- Capability inventory: Subprocess execution is limited to calling the
aliyunbinary for cloud management tasks. - Sanitization: While explicit string sanitization is not detailed, the skill focuses on schema-driven orchestration where external help text is used for discovery rather than direct prompt interpolation for logic control.
Audit Metadata