The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent on how to use lark-cli for authentication, configuration, and resource management. It specifically guides the agent to use argument arrays for command execution to prevent shell injection vulnerabilities.
[EXTERNAL_DOWNLOADS]: Includes instructions for the lark-cli update command, which fetches the latest versions of the CLI and its associated skills from the vendor's sources.
[SAFE]: Implements a 'high-risk operation protocol' that uses specific exit codes (exit 10) and JSON responses to force the agent to request explicit user confirmation before proceeding with destructive actions like deletions.
[SAFE]: Explicitly prohibits the output of sensitive information such as appSecret or accessToken to the terminal or agent responses.

lark-shared