The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides instructions to download and execute a setup script directly into the shell for the Alibaba Cloud CLI.
Evidence: The command curl -fsSL https://aliyuncli.alicdn.com/setup.sh | bash is recommended in SKILL.md to update the CLI tool. The source URL aliyuncli.alicdn.com is an official vendor resource for Alibaba Cloud.
[INDIRECT_PROMPT_INJECTION]: The skill includes capabilities to read external data, specifically job logs from Alibaba Cloud OSS storage, which could contain untrusted content.
Ingestion points: references/api-reference.md and references/job-management.md detail methods to fetch log content (e.g., stdout.log, stderr.log) from OSS buckets via the list-log-contents command.
Boundary markers: The skill does not provide specific delimiters or instructions to the agent to disregard potential instructions embedded within the retrieved log data.
Capability inventory: The agent has the capability to execute a wide range of Alibaba Cloud CLI commands, including resource creation, modification, and RAM policy management.
Sanitization: No sanitization or validation steps are documented for the retrieved log content before the agent processes or interprets it.
[COMMAND_EXECUTION]: The skill's primary function is the execution of system shell commands via the aliyun CLI tool to manage cloud infrastructure and Spark workloads.
[PRIVILEGE_ESCALATION]: The skill contains instructions for managing Identity and Access Management (RAM) policies, which involves granting permissions to users.
Evidence: references/ram-policies.md provides commands like aliyun ram attach-policy-to-user to grant administrative or developer permissions. While these are administrative functions inherent to the skill's purpose, they represent highly privileged operations.

alibabacloud-emr-spark-manage