The Agent Skills Directory

[SAFE]: The skill consists entirely of Markdown-based guidance, templates, and decision trees for architectural analysis. No scripts, binaries, or executable logic are present.
[SAFE]: The configuration in SKILL.md limits the agent to the Read, Glob, and Grep tools, effectively enforcing a read-only policy that prevents unauthorized file modifications or network exfiltration.
[SAFE]: All referenced 'Related Skills' follow standard internal linking patterns, and there are no external dependencies or remote script invocations.

architecture