## Core Rule

<!--
One sentence (max ~25 words) that anchors the skill's ethical scope:
- What must be preserved
- What's forbidden even on user request
- The deal-breaker that distinguishes this skill from a free-form prompt

Examples:
- "Confirm code is unreachable across the full call graph before flagging. Never remove code without explicit user approval, even for obvious dead branches."
- "Block deployment when CHANGELOG, tests, or coverage gates fail. Preserve rollback capability at every step. Never ship from a dirty working tree."
- "Form a single hypothesis at a time. Validate with a falsifiable test before fixing. Never patch a symptom you haven't reproduced."

Required for every skill. Goes immediately after the title (and intro paragraph if present), before any other ## section.
-->

[Replace with this skill's Core Rule.]