alibabacloud-resourcecenter-search
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads the Alibaba Cloud CLI installation script and binaries from
aliyuncli.alicdn.com. As the skill is authored by 'aliyun', these represent official vendor resources used to satisfy prerequisites. - [COMMAND_EXECUTION]: The Python script
scripts/query-resource-types.pyutilizessubprocess.run()to execute thealiyunCLI. The command parameters are largely static, and user-provided keywords are processed within Python logic rather than being passed directly to the shell, effectively mitigating command injection risks. - [INDIRECT_PROMPT_INJECTION]: The skill processes data fetched from Alibaba Cloud APIs. While this constitutes an attack surface where untrusted data could theoretically contain instructions, the risk is minimal given the structured nature of the resource metadata and the lack of high-privilege autonomous write capabilities without confirmation.
- [CREDENTIALS_UNSAFE]: The skill instructions proactively include security rules that forbid the agent from reading, echoing, or asking the user for Access Key/Secret Key (AK/SK) values. It correctly directs users to configure credentials via the CLI outside of the active session.
Audit Metadata