The Agent Skills Directory

[SAFE]: The skill is comprised entirely of Markdown-based instructions, templates, and evaluation examples. There are no executable scripts (.sh, .py, .js), binaries, or automated commands included in the package.
[DATA_EXPOSURE]: The skill includes defensive instructions in SKILL.md and references/RUBRIC.md that explicitly direct users and the agent to avoid requesting or providing PII (Personally Identifiable Information) or secrets, favoring redacted or aggregated data instead.
[COMMAND_EXECUTION]: No shell command execution patterns or dynamic context injections (such as the ! syntax) were found in any of the skill's files.
[EXTERNAL_DOWNLOADS]: There are no automated remote code execution patterns, package installations (npm/pip), or network-based data fetching mechanisms. External URLs are provided for metadata and informational reference only.

user-onboarding