The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes lark-cli commands for authentication (auth login, auth status), configuration (config init), and identity management (whoami). These are standard operations for a CLI-based agent skill.
[CREDENTIALS_SAFE]: The skill explicitly instructs the agent to avoid outputting secrets like appSecret or accessToken to the terminal.
[PRIVILEGE_ESCALATION]: The skill implements a high-risk confirmation protocol (exit code 10) that requires explicit user consent before proceeding with dangerous write or delete operations.
[EXTERNAL_DOWNLOADS]: The skill references official Lark/Feishu documentation and tools from the larksuite GitHub organization, which is a trusted vendor.

lark-shared