The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes lifecycle management commands for cloud sandboxes using the bl CLI tool.- [CREDENTIALS_UNSAFE]: The skill provides clear instructions for the secure handling of API keys and session tokens, including the use of restricted-permission temporary directories and redaction of secrets in user-facing output.- [DYNAMIC_EXECUTION]: The skill contains Python and Node.js code snippets designed to be run by the agent to interact with the sandbox runtime (WebShell and CDP). These are standard operational templates.- [EXTERNAL_DOWNLOADS]: The skill references the puppeteer-core Node.js package for browser automation tasks. This is a well-known and legitimate dependency for the described functionality.

bailian-sandbox