v0.2.0

Added

• New Core Rule "Distinguish verified from believed" (skills/be-precise/SKILL.md). Closes the gap where a checkable-but-unverified inference (a cause, a mechanism, how an API behaves) gets stated as fact: state as fact only what you checked against the source, label or verify everything else, and when challenged re-verify at the source rather than defend an unconfirmed position.
• Two Common Mistakes entries mirroring the rule as failure modes: "Stating an unverified inference as fact" and "Defending a challenged claim instead of re-checking".