alicloud-ai-recommend-airec
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFEDATA_EXFILTRATIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill instructions and configuration specify accessing Alibaba Cloud credentials from standard environment variables (ALICLOUD_ACCESS_KEY_ID, ALICLOUD_ACCESS_KEY_SECRET) and the shared local credentials file at ~/.alibabacloud/credentials. This access is required for the skill's primary function of cloud resource management and is documented clearly.
- [EXTERNAL_DOWNLOADS]: The script scripts/list_openapi_meta_apis.py fetches OpenAPI documentation from api.aliyun.com. This is the official metadata endpoint for Alibaba Cloud services and is considered a well-known, trusted source.
- [COMMAND_EXECUTION]: The skill executes its internal python scripts for metadata discovery and uses the standard py_compile module for syntax validation during its setup phase. No arbitrary or high-risk command execution patterns were found.
- [PROMPT_INJECTION]: The skill processes external API metadata which represents a surface for indirect prompt injection.
- Ingestion points: scripts/list_openapi_meta_apis.py (line 40) fetches JSON data from api.aliyun.com.
- Boundary markers: The skill does not currently implement explicit delimiters or instruction-ignoring directives for the external metadata.
- Capability inventory: The skill has the ability to write files to the local output directory and perform network requests to official cloud endpoints.
- Sanitization: Data is parsed using standard JSON libraries, but the content is not validated against a strict schema to prevent instruction injection.
Audit Metadata