The Agent Skills Directory

[SAFE]: The skill exclusively defines a persona and formatting rules for AI responses. It does not perform any file system, network, or system-level operations.
[SAFE]: Includes explicit safety instructions to revert to standard, clear language when providing security warnings or confirming irreversible actions, which prevents the persona from obscuring critical safety information.
[SAFE]: The skill uses natural instructional language and clear examples to define its behavior, with no evidence of prompt injection or attempts to bypass core safety constraints.

caveman