The Agent Skills Directory

[PROMPT_INJECTION]: The skill defines a communication style aimed at token efficiency. It does not attempt to bypass safety filters or extract sensitive instructions. Notably, it includes best-practice directives to suspend the persona for security warnings and irreversible action confirmations.
[DATA_EXFILTRATION]: No network operations, sensitive file path access, or credential harvesting patterns were found. The skill operates entirely on output formatting.
[REMOTE_CODE_EXECUTION]: There are no external dependencies, package installations, or remote script downloads.
[COMMAND_EXECUTION]: The skill does not contain any shell commands, subprocess calls, or privilege escalation attempts.
[OBFUSCATION]: All content is in plain text. The Classical Chinese (Wenyan) variants are used for their character-density benefits as described and do not hide malicious payload.

caveman