The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions in references/cli-installation-guide.md to download the Aliyun CLI binary and setup scripts from aliyuncli.alicdn.com, which is the vendor's official distribution domain.
[REMOTE_CODE_EXECUTION]: The references/cli-installation-guide.md file contains a command curl -fsSL https://aliyuncli.alicdn.com/setup.sh | bash which downloads and executes a remote script directly. This is a standard installation method provided by the vendor.
[COMMAND_EXECUTION]: The skill's primary function is to construct and execute shell commands using the aliyun CLI tool to manage cloud resources. This includes actions like create-dataset, delete-dataset, and execute-query.
[PRIVILEGE_ESCALATION]: The installation guide recommends using sudo mv to place the aliyun binary into /usr/local/bin/, which requires elevated privileges on the host system.
[INDIRECT_PROMPT_INJECTION]: The skill represents a vulnerability surface for indirect prompt injection as it interpolates user-controlled data—such as workspace IDs, dataset names, and SQL queries—directly into shell commands without explicit sanitization mechanisms beyond a general mention of 'safe shell quoting'.
Ingestion points: User-provided parameters for workspace, dataset-name, description, schema, and query in SKILL.md.
Boundary markers: The instructions use single quotes for JSON and queries to mitigate shell interpolation issues, but do not employ robust prompt delimiters for external data.
Capability inventory: The skill has the capability to execute arbitrary commands via the aliyun CLI, including data querying and resource modification.
Sanitization: There is no evidence of automated validation or escaping of the user-provided strings before they are passed to the CLI.

alibabacloud-cms-dataset