The Agent Skills Directory

[SAFE]: The skill instructions and metadata do not contain any malicious patterns or security risks. All commands are restricted to the lark-cli tool.
[PROMPT_INJECTION]: No evidence of instructions intended to bypass safety guardrails, override system prompts, or reveal internal configurations. The use of 'CRITICAL' in the documentation is for instructional workflow (modular skill loading) rather than a safety bypass.
[DATA_EXFILTRATION]: No hardcoded credentials, sensitive file access, or unauthorized network operations were identified. The skill points to a shared configuration file for authentication handling.
[COMMAND_EXECUTION]: The skill utilizes a specific CLI tool (lark-cli) for its intended purpose. There is no evidence of arbitrary shell execution or unsafe user input interpolation.

lark-approval