The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data, such as domain names and filing organizer details, which are used as arguments for CLI commands (aliyun domain ...). This creates a surface where malicious input could potentially influence cloud management tasks.\n
Ingestion points: User-provided domain names, ICP filing data, and messages for AI-driven website generation.\n
Boundary markers: Instructions state that all user-provided identifiers and command output must be treated as untrusted data, though no technical delimiters are enforced in the prompts.\n
Capability inventory: The skill possesses significant capabilities including domain registration, DNS record modification, and website deployment using the aliyun CLI tool.\n
Sanitization: Relies on internal CLI validation and the requirement to verify target identifiers against successful discovery results before use.\n- [COMMAND_EXECUTION]: The skill is designed to execute the aliyun Alibaba Cloud CLI binary. It includes a local Python preflight script (scripts/eval_domain_preflight.py) that uses subprocess.run to verify CLI versioning and environment readiness.\n- [EXTERNAL_DOWNLOADS]: The documentation mentions the official Alibaba Cloud CLI setup script (https://aliyuncli.alicdn.com/setup.sh). This reference targets the vendor's official distribution infrastructure and the instructions explicitly prohibit the skill from implicitly executing the installer.

alibabacloud-domain