Install any skill in seconds. Free to start, no credit card required.
Get Started Free →融合《掌控习惯》(James Clear) 四大定律与《福格行为模型》(B.J. Fogg) B=MAP 的行为设计技能。适用于用户要养成新习惯或戒除坏习惯、诊断"计划总是坚持不下去"的问题、为产品或团队设计行为改变机制、以及把大目标拆解为可持续微行动时使用。
.claude/skills/kuhung-behavior-design/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 90% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 52% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 44% | 0% |
| case-01 | ✓→✓ | = Same ✓ | 24% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 53% | 0% |
你是一名行为设计教练,整合两套互补的行为科学框架:福格的 B=MAP 模型(行为发生的机制)与克利尔的四大定律(习惯固化的工程)。你的使命是帮助用户停止依赖意志力和动机,改用设计手段让好行为自然发生。
详细论据分别见 notes/掌控习惯_笔记.md 与 notes/福格行为模型_笔记.md。
暂无。技能在实战中暴露的偏差会以 - YYYY-MM-DD: 经验内容 格式追加到本章节。
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 14,952 | 12,282 | -18% | 1 | 1 | 0% | 2,419 | 2,990 | +24% | 0 | 0 | — |
case-02 | fail→pass | 10,365 | 13,968 | +35% | 1 | 1 | 0% | 1,677 | 3,187 | +90% | 0 | 0 | — |
case-03 | fail→pass | 11,852 | 11,422 | -4% | 1 | 1 | 0% | 1,893 | 2,872 | +52% | 0 | 0 | — |
case-04 | pass→pass | 10,505 | 10,539 | +0% | 1 | 1 | 0% | 1,790 | 2,747 | +53% | 0 | 0 | — |
case-05 | pass→pass | 11,682 | 10,789 | -8% | 1 | 1 | 0% | 2,189 | 2,794 | +28% | 0 | 0 | — |
case-06 | pass→pass | 11,919 | 12,628 | +6% | 1 | 1 | 0% | 2,298 | 3,002 | +31% | 0 | 0 | — |
case-07 | pass→pass | 10,992 | 12,085 | +10% | 1 | 1 | 0% | 1,774 | 2,903 | +64% | 0 | 0 | — |
case-08 | pass→pass | 14,540 | 13,679 | -6% | 1 | 1 | 0% | 2,187 | 3,054 | +40% | 0 | 0 | — |
case-09 | pass→pass | 15,528 | 12,539 | -19% | 1 | 1 | 0% | 1,534 | 2,909 | +90% | 0 | 0 | — |
case-10 | pass→pass | 11,205 | 9,741 | -13% | 1 | 1 | 0% | 1,799 | 2,568 | +43% | 0 | 0 | — |
case-11 | fail→pass | 13,279 | 12,781 | -4% | 1 | 1 | 0% | 2,090 | 3,002 | +44% | 0 | 0 | — |
case-12 | pass→pass | 11,302 | 11,521 | +2% | 1 | 1 | 0% | 1,642 | 2,839 | +73% | 0 | 0 | — |
case-13 | pass→pass | 9,617 | 11,483 | +19% | 1 | 1 | 0% | 1,519 | 2,767 | +82% | 0 | 0 | — |
case-14 | fail→fail | 12,802 | 14,569 | +14% | 1 | 1 | 0% | 1,984 | 3,265 | +65% | 0 | 0 | — |
case-15 | pass→pass | 9,774 | 9,552 | -2% | 1 | 1 | 0% | 1,632 | 2,631 | +61% | 0 | 0 | — |
case-16 | pass→pass | 10,796 | 11,677 | +8% | 1 | 1 | 0% | 1,621 | 2,819 | +74% | 0 | 0 | — |
case-17 | pass→pass | 11,711 | 13,301 | +14% | 1 | 1 | 0% | 1,844 | 2,964 | +61% | 0 | 0 | — |
case-18 | pass→pass | 11,099 | 13,114 | +18% | 1 | 1 | 0% | 1,729 | 3,165 | +83% | 0 | 0 | — |
case-19 | pass→pass | 10,179 | 12,187 | +20% | 1 | 1 | 0% | 1,635 | 2,840 | +74% | 0 | 0 | — |
case-20 | pass→pass | 10,331 | 12,954 | +25% | 1 | 1 | 0% | 1,731 | 3,033 | +75% | 0 | 0 | — |
case-21 | fail→fail | 12,410 | 12,664 | +2% | 1 | 1 | 0% | 2,028 | 2,971 | +46% | 0 | 0 | — |
case-22 | pass→pass | 11,700 | 12,528 | +7% | 1 | 1 | 0% | 1,782 | 3,107 | +74% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +14 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.