alicloud-platform-devops

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFEDATA_EXFILTRATIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill is designed to access sensitive Alibaba Cloud credentials from environment variables such as ALICLOUD_ACCESS_KEY_ID and ALICLOUD_ACCESS_KEY_SECRET, as well as the standard local credentials file located at ~/.alibabacloud/credentials.
  • [EXTERNAL_DOWNLOADS]: Fetches machine-readable OpenAPI metadata and documentation from official Alibaba Cloud domains (api.aliyun.com) to provide up-to-date API discovery features.
  • [COMMAND_EXECUTION]: Includes multiple Python scripts (e.g., list_projects.py, list_repositories.py) that execute locally to perform resource inventory and inspection tasks via the Alibaba Cloud SDK.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by ingesting external API metadata.
  • Ingestion points: The scripts/list_openapi_meta_apis.py file downloads JSON-formatted API documentation from an external URL.
  • Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious instructions embedded within the fetched metadata.
  • Capability inventory: The skill has the ability to write files to the local output directory and perform authenticated API operations against Alibaba Cloud services.
  • Sanitization: No explicit sanitization or structural validation is performed on the ingested JSON data before it is processed and presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 10:15 AM