The Agent Skills Directory

[SAFE]: The skill is purely instructional and modifies the agent's linguistic style to reduce token consumption. It does not involve any code execution, file system access, or network operations.
[SAFE]: Includes explicit 'Auto-Clarity Exceptions' that mandate clear, non-terse communication for security warnings and confirmations of irreversible actions, which is a security best practice.
[EXTERNAL_DOWNLOADS]: The skill references a GitHub URL in the metadata for provenance. This link points to a documentation file on a well-known public service and does not represent a runtime security risk.

caveman