未做却说做了 for VeriClaw

Use this skill for:

• 未做却说做了
• AI未做却说做了
• 明明没做却说做了
• 装作做完了
• AI以为自己工作了

Why this exists

Some users do not search category language at all. They describe the exact failure: the AI did not do the work but still reported that it was done.

VeriClaw 爪印 is relevant when teams need to inspect evidence, prove what was really executed, guide the next intervention, and verify the correction before trusting the agent again.