alibabacloud-agentloop-contextstore
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references and downloads the Aliyun CLI installation script and binary packages from aliyuncli.alicdn.com. This domain is the official CDN for Alibaba Cloud's command-line tools.
- [REMOTE_CODE_EXECUTION]: Instructions in the installation guide include a command to download and execute a setup script via a shell pipe (curl ... | bash). This is a standard and documented method for installing the vendor's official tool.
- [COMMAND_EXECUTION]: The skill's primary function is to execute aliyun CLI commands to manage cloud resources. It includes instructions to enable AI-mode and configure the User-Agent for tracking, which are standard features of the Aliyun CLI for agentic workflows.
- [CREDENTIALS_UNSAFE]: The skill provides robust guidance on credential safety, explicitly warning users and the agent never to read, echo, or print access key secrets. It recommends using aliyun configure list to verify authentication status without exposing sensitive values.
- [PROMPT_INJECTION]: The skill processes user-provided data for context storage using the aliyun CLI. Ingestion points: Command arguments in SKILL.md. Boundary markers: Absent. Capability inventory: Subprocess calls to aliyun CLI in references/store-management.md. Sanitization: Relies on Aliyun CLI's internal JSON parsing and service-side validation.
Audit Metadata