The Agent Skills Directory

[SAFE]: The skill establishes a structured process for analyzing engineering requests, including reality checks, effort classification, and alternative path exploration.
[SAFE]: References to well-known services (Stripe, Auth0, Twilio, SendGrid) and package managers (npm, pip, crates, gems) are used as examples for architectural research and do not involve credential exposure or network exfiltration.
[SAFE]: The skill instructs the agent to use platform-specific tools like reference-engine and github-search for information gathering, which is standard operational behavior.
[NO_CODE]: The skill consists of instructional content and workflow diagrams without any executable code, scripts, or binary assets.

rationale