Health checks for AI agent systems that also answer 'could I even tell?'. Unmeasurable never counts as green. One file, no dependencies.
-
Updated
Sep 2, 2026 - Python
Health checks for AI agent systems that also answer 'could I even tell?'. Unmeasurable never counts as green. One file, no dependencies.
A Claude Code metaskill that enforces a mandatory self-test → reflect → rework → quality-gate closed loop before any delivery, making the model accountable for its own output quality.
UI/UX design makeover skill for Claude Code. Audits your app, proposes complete directions, builds, and self-tests before you see the result. Stops AI-generic UI cold.
Practical shortcuts and science-backed techniques for learning German faster and remembering longer.
To associate your repository with the self-testing topic, visit your repo's landing page and select "manage topics."