The Agent Skills Directory

[SAFE]: The skill's primary function is analysis and recommendation. It follows a structured workflow (QUESTION → WEIGH → SUBTRACT → PROPOSE) to evaluate the necessity of project artifacts without performing any execution.
[SAFE]: No sensitive data exposure or exfiltration risks were identified. There are no network-related commands (curl, wget, etc.) or access to sensitive file paths.
[SAFE]: The skill explicitly defines strict boundaries, stating it must 'Never edit code or documents directly' and 'Never execute deletion or refactoring work directly,' instead routing implementation to other agents.
[SAFE]: No obfuscation, prompt injection, or remote code execution vectors are present. External links provided in the reference documentation point to legitimate industry-standard technical resources (e.g., SonarSource, Martin Fowler's blog, Amazon, and HBR).

void