The Agent Skills Directory

[SAFE]: The skill is designed for code exploration and architectural explanation. It uses a structured multi-agent approach (explorers and explainers) to synthesize findings from a codebase.
[SAFE]: In SKILL.md, the instructions explicitly state: "Every subagent here is read-only: tell it not to edit files." This is a positive safety constraint.
[SAFE]: The templates in references/explainer-prompt.md and references/explorer-prompt.md provide clear, professional guidelines for subagents to follow, focusing on senior-engineer-level architectural insights.
[SAFE]: No network operations, credential access, or persistent mechanisms were detected. The skill relies on standard platform tools like Grep, Glob, and Read for analysis.

how