Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Support a teenager through college applications without taking them over — the parent's actual jobs (logistics, finances, emotional ballast), the ownership lines that keep the application theirs, and the scripts for the hard moments. Use when asked how do I help my kid with college apps, how involved should I be, my teenager won't start their essays, or we disagree about the college list. Produces the role split, the family timeline, the money conversation framework, and the scripts for deadlock
.claude/skills/mohitagw15856-college-app-parent-guide/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 25% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 67% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 51% | 0% |
The college application is a project with two failure modes at home: the parent who runs it (producing an application that got in somewhere the student never chose, written in a voice admissions readers recognize as forty-five years old) and the parent who vanishes (leaving a seventeen-year-old to solo-manage a dozen deadlines and a five-figure decision). The working role is specific: the parent owns logistics, money truth, and emotional ballast; the student owns the list, the essays, and the clicks. This skill draws those lines, builds the shared timeline, and scripts the moments where the lines get tested.
Ask for these if not provided:
Yours: the logistics/money/ballast list] · Theirs: the list/essays/asks list] · The rule: you may ask about the calendar; you don't touch the keyboard.
Backward from deadlines: tests, visits, asks, drafts, submissions · the weekly 20-minute check-in that replaces daily nagging]
The contribution number and how to say it · net-price homework to do together · the loan-reality sentence]
Stalled: "small scope + company]" · List fight: "input once + money veto + their call]" · Essay ask: "the react-don't-write line]" · Rejection day: "hour-one script]"
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 20,828 | 15,150 | -27% | 1 | 1 | 0% | 3,306 | 4,124 | +25% | 0 | 0 | — |
case-02 | fail→fail | 16,756 | 20,206 | +21% | 1 | 1 | 0% | 3,033 | 4,686 | +55% | 0 | 0 | — |
case-03 | fail→pass | 21,688 | 14,411 | -34% | 1 | 1 | 0% | 3,754 | 3,734 | -1% | 0 | 0 | — |
case-04 | fail→fail | 16,180 | 14,063 | -13% | 1 | 1 | 0% | 3,126 | 3,567 | +14% | 0 | 0 | — |
case-05 | pass→pass | 12,640 | 11,743 | -7% | 1 | 1 | 0% | 1,920 | 3,321 | +73% | 0 | 0 | — |
case-06 | pass→pass | 12,569 | 12,523 | -0% | 1 | 1 | 0% | 2,180 | 3,430 | +57% | 0 | 0 | — |
case-07 | fail→pass | 12,737 | 14,370 | +13% | 1 | 1 | 0% | 2,177 | 3,635 | +67% | 0 | 0 | — |
case-08 | fail→pass | 10,343 | 9,479 | -8% | 1 | 1 | 0% | 1,806 | 2,786 | +54% | 0 | 0 | — |
case-09 | fail→pass | 14,109 | 12,128 | -14% | 1 | 1 | 0% | 2,110 | 3,176 | +51% | 0 | 0 | — |
case-10 | pass→pass | 11,821 | 12,185 | +3% | 1 | 1 | 0% | 2,048 | 3,235 | +58% | 0 | 0 | — |
case-11 | pass→pass | 13,204 | 12,050 | -9% | 1 | 1 | 0% | 2,321 | 3,275 | +41% | 0 | 0 | — |
case-12 | pass→pass | 11,897 | 9,906 | -17% | 1 | 1 | 0% | 1,886 | 2,842 | +51% | 0 | 0 | — |
case-18 | pass→pass | 14,092 | 12,394 | -12% | 1 | 1 | 0% | 2,321 | 3,284 | +41% | 0 | 0 | — |
case-13 | fail→pass | 11,693 | 11,667 | -0% | 1 | 1 | 0% | 1,994 | 3,199 | +60% | 0 | 0 | — |
case-14 | fail→pass | 16,091 | 13,031 | -19% | 1 | 1 | 0% | 2,672 | 3,457 | +29% | 0 | 0 | — |
case-15 | pass→pass | 10,526 | 9,199 | -13% | 1 | 1 | 0% | 1,836 | 2,833 | +54% | 0 | 0 | — |
case-16 | pass→pass | 7,822 | 5,441 | -30% | 1 | 1 | 0% | 1,382 | 2,238 | +62% | 0 | 0 | — |
case-17 | pass→pass | 11,422 | 11,685 | +2% | 1 | 1 | 0% | 1,808 | 3,116 | +72% | 0 | 0 | — |
case-19 | pass→pass | 13,758 | 8,882 | -35% | 1 | 1 | 0% | 2,219 | 2,701 | +22% | 0 | 0 | — |
case-20 | pass→pass | 13,425 | 11,461 | -15% | 1 | 1 | 0% | 2,213 | 3,058 | +38% | 0 | 0 | — |
case-21 | pass→pass | 14,625 | 12,115 | -17% | 1 | 1 | 0% | 2,821 | 3,505 | +24% | 0 | 0 | — |
case-22 | pass→pass | 15,632 | 13,881 | -11% | 1 | 1 | 0% | 2,589 | 3,711 | +43% | 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 +32 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.