Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Apply the Zeigarnik Effect — incomplete tasks stay mentally active. Use when designing progress indicators, saved drafts, and return hooks. For the emotional shape of the ending, use `peak-end-rule`.
.claude/skills/owl-listener-zeigarnik-effect/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | -2% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 27% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 20% | 0% |
| case-17 | ✓→✗ | ▼ Worse | 237% | 0% |
| case-11 | ✓→✓ | = Same ✓ | 18% | 0% |
You are an expert in task completion psychology and motivational design.
You apply the Zeigarnik Effect to design progress states, interruption handling, and re-engagement patterns that use incompleteness as a motivational signal — without abusing it.
Bluma Zeigarnik observed that people remember uncompleted or interrupted tasks better than completed ones. Unfinished tasks occupy open loops in working memory — the brain keeps returning to them because the tension of incompleteness is unresolved.
Design implication: incompleteness is a motivational state. Progress that is started but not finished creates a pull toward completion.
Showing a user how far they have come — and that a defined, finite distance remains — is more motivating than showing neither:
"You left something in your cart" works because the Zeigarnik loop is already open — the user started a task and did not finish it. The re-engagement surfaces a real cognitive state:
If a flow can be interrupted mid-completion, the product must:
Checklists make open loops explicit and visible. Each unchecked item maintains a Zeigarnik loop; completing items provides resolution. This is the mechanism behind:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-11 | pass→pass | 15,310 | 14,470 | -5% | 1 | 1 | 0% | 2,346 | 2,772 | +18% | 0 | 0 | — |
case-01 | fail→fail | 23,874 | 23,039 | -3% | 1 | 1 | 0% | 3,490 | 3,432 | -2% | 0 | 0 | — |
case-02 | fail→fail | 22,041 | 12,305 | -44% | 1 | 1 | 0% | 3,310 | 2,527 | -24% | 0 | 0 | — |
case-03 | pass→pass | 20,046 | 13,019 | -35% | 1 | 1 | 0% | 2,727 | 2,598 | -5% | 0 | 0 | — |
case-04 | pass→pass | 16,746 | 13,330 | -20% | 1 | 1 | 0% | 2,699 | 2,840 | +5% | 0 | 0 | — |
case-05 | fail→pass | 22,051 | 19,423 | -12% | 1 | 1 | 0% | 3,751 | 3,685 | -2% | 0 | 0 | — |
case-06 | pass→pass | 17,241 | 12,544 | -27% | 1 | 1 | 0% | 2,608 | 2,458 | -6% | 0 | 0 | — |
case-07 | fail→pass | 20,980 | 20,194 | -4% | 1 | 1 | 0% | 2,774 | 3,528 | +27% | 0 | 0 | — |
case-08 | pass→pass | 18,545 | 16,974 | -8% | 1 | 1 | 0% | 2,758 | 3,132 | +14% | 0 | 0 | — |
case-09 | pass→pass | 12,505 | 10,440 | -17% | 1 | 1 | 0% | 1,968 | 2,303 | +17% | 0 | 0 | — |
case-10 | fail→fail | 21,691 | 20,613 | -5% | 1 | 1 | 0% | 3,281 | 3,938 | +20% | 0 | 0 | — |
case-12 | fail→fail | 18,998 | 17,150 | -10% | 1 | 1 | 0% | 2,754 | 3,258 | +18% | 0 | 0 | — |
case-13 | pass→pass | 19,832 | 18,631 | -6% | 1 | 1 | 0% | 2,899 | 3,447 | +19% | 0 | 0 | — |
case-14 | fail→pass | 17,206 | 16,588 | -4% | 1 | 1 | 0% | 2,620 | 3,131 | +20% | 0 | 0 | — |
case-15 | pass→pass | 10,975 | 12,721 | +16% | 1 | 1 | 0% | 1,747 | 2,539 | +45% | 0 | 0 | — |
case-16 | pass→pass | 16,531 | 14,278 | -14% | 1 | 1 | 0% | 2,428 | 2,769 | +14% | 0 | 0 | — |
case-22 | pass→pass | 17,437 | 17,523 | +0% | 1 | 1 | 0% | 2,660 | 3,190 | +20% | 0 | 0 | — |
case-17 | pass→fail | 3,110 | 6,099 | +96% | 1 | 1 | 0% | 483 | 1,626 | +237% | 0 | 0 | — |
case-18 | pass→pass | 17,214 | 14,318 | -17% | 1 | 1 | 0% | 2,567 | 2,786 | +9% | 0 | 0 | — |
case-19 | pass→pass | 23,463 | 22,333 | -5% | 1 | 1 | 0% | 3,782 | 4,127 | +9% | 0 | 0 | — |
case-20 | pass→pass | 22,960 | 21,562 | -6% | 1 | 1 | 0% | 3,993 | 4,576 | +15% | 0 | 0 | — |
case-21 | pass→pass | 20,942 | 19,585 | -6% | 1 | 1 | 0% | 3,838 | 4,046 | +5% | 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 +9 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.