alibabacloud-idaas-cloud-role-management

Pass

Audited by Gen Agent Trust Hub on Oct 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes cloud management tasks by invoking the official aliyun CLI and other provider CLIs. These calls are performed using subprocess.run with shell=False, preventing shell injection vulnerabilities. The arguments for these commands are strictly validated against pre-approved schemas.
  • [EXTERNAL_DOWNLOADS]: The skill can update or install official CLI plugins from Alibaba Cloud's official distribution server (https://aliyuncli.alicdn.com/plugins). This functionality is limited to the official vendor domain and is used only to maintain the local execution environment when authorized by the --repair flag.
  • [INDIRECT_PROMPT_INJECTION]: The skill interprets user-provided permission requirements to generate cloud policies. It mitigates the risk of malicious instructions within these requirements by generating a 'sealed plan' containing the exact policy text and a SHA256 hash. The agent is strictly instructed to display this plan and wait for explicit user confirmation before any mutations are performed.
  • [CREDENTIALS_UNSAFE]: The bundled scripts/role_tools.py script contains a scanning mechanism that identifies and rejects common sensitive keys (such as access_key, secret, password, token) within role request artifacts, preventing the accidental ingestion or processing of hardcoded credentials.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 29, 2026, 07:36 AM