The Agent Skills Directory

[SAFE]: The skill content is purely instructional, defining frameworks for ADRs and implementation planning without any malicious code, command execution, or network requests.
[SAFE]: Allowed tools are restricted to Read, Grep, and Glob, ensuring the skill only accesses codebase metadata and content without the ability to modify files or exfiltrate data.
[SAFE]: The skill explicitly promotes security-conscious behavior by recommending the use of pre-commit and pre-push hooks for secret detection, formatting, and test coverage validation.

architecture-patterns