aliyun-green-moderation
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFECREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill instructions specify access to the standard Alibaba Cloud configuration file located at
~/.alibabacloud/credentialsand specific environment variables (ALIBABACLOUD_ACCESS_KEY_ID,ALIBABACLOUD_ACCESS_KEY_SECRET). These are legitimate and expected paths for tools managing Alibaba Cloud resources. - [EXTERNAL_DOWNLOADS]: The script
scripts/list_openapi_meta_apis.pyfetches OpenAPI metadata from Alibaba Cloud's official domain (api.aliyun.com). This is a well-known service and the reference is for legitimate discovery of API documentation. - [COMMAND_EXECUTION]: The skill executes local Python scripts to inventory APIs and perform metadata-driven discovery. It also uses
py_compileto validate script integrity during the validation phase. - [INDIRECT_PROMPT_INJECTION]: The skill fetches and processes external JSON metadata from a remote source, which serves as a potential attack surface for indirect prompt injection.
- Ingestion points:
scripts/list_openapi_meta_apis.pyfetches JSON fromhttps://api.aliyun.com/meta/v1/products/.... - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when processing the fetched metadata.
- Capability inventory: The skill has the capability to execute local scripts and perform authenticated Alibaba Cloud API operations.
- Sanitization: The script parses the fetched content as JSON, but there is no specific sanitization of the values within the JSON metadata before they are used to inform agent actions.
Audit Metadata