Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Guides borrowers through mortgage refinance evaluation — collects loan data, extracts mortgage statement fields, evaluates qualification, and delivers recommendations with consumer-friendly communication.
.claude/skills/bilal140202-mortgage-loan-officer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -78% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -75% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 0% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 17% | 0% |
Knowledgeable mortgage loan officer assistant that guides borrowers through a refinance evaluation by collecting information, extracting data from mortgage statements, analyzing refinance scenarios, and delivering clear recommendations.
This skill is part of the mortgage plugin. Install via:
/plugin marketplace add lendtrain/mortgage
/plugin install mortgage@mortgageFull source: github.com/lendtrain/mortgage
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 10,588 | 8,603 | -19% | 1 | 1 | 0% | 1,891 | 1,701 | -10% | 0 | 0 | — |
case-02 | fail→pass | 12,077 | 12,666 | +5% | 1 | 1 | 0% | 2,284 | 2,519 | +10% | 0 | 0 | — |
case-03 | fail→fail | 13,173 | 12,501 | -5% | 1 | 1 | 0% | 2,248 | 2,413 | +7% | 0 | 0 | — |
case-04 | pass→pass | 10,025 | 7,441 | -26% | 1 | 1 | 0% | 1,564 | 1,557 | -0% | 0 | 0 | — |
case-05 | pass→pass | 6,936 | 6,939 | +0% | 1 | 1 | 0% | 1,223 | 1,431 | +17% | 0 | 0 | — |
case-06 | pass→pass | 5,938 | 3,828 | -36% | 1 | 1 | 0% | 896 | 848 | -5% | 0 | 0 | — |
case-07 | pass→pass | 3,320 | 3,659 | +10% | 1 | 1 | 0% | 478 | 843 | +76% | 0 | 0 | — |
case-08 | pass→pass | 8,911 | 7,224 | -19% | 1 | 1 | 0% | 1,562 | 1,578 | +1% | 0 | 0 | — |
case-09 | pass→pass | 7,088 | 6,169 | -13% | 1 | 1 | 0% | 1,259 | 1,361 | +8% | 0 | 0 | — |
case-10 | pass→pass | 8,770 | 10,938 | +25% | 1 | 1 | 0% | 1,617 | 2,216 | +37% | 0 | 0 | — |
case-11 | pass→pass | 5,854 | 5,346 | -9% | 1 | 1 | 0% | 1,019 | 1,244 | +22% | 0 | 0 | — |
case-12 | fail→pass | 13,267 | 1,894 | -86% | 1 | 1 | 0% | 2,292 | 500 | -78% | 0 | 0 | — |
case-13 | fail→pass | 12,432 | 1,638 | -87% | 1 | 1 | 0% | 2,080 | 526 | -75% | 0 | 0 | — |
case-14 | pass→pass | 14,758 | 7,078 | -52% | 1 | 1 | 0% | 2,680 | 1,577 | -41% | 0 | 0 | — |
case-15 | pass→pass | 16,580 | 13,723 | -17% | 1 | 1 | 0% | 2,926 | 2,762 | -6% | 0 | 0 | — |
case-16 | pass→pass | 4,334 | 4,369 | +1% | 1 | 1 | 0% | 833 | 1,042 | +25% | 0 | 0 | — |
case-17 | pass→pass | 5,334 | 3,189 | -40% | 1 | 1 | 0% | 926 | 684 | -26% | 0 | 0 | — |
case-18 | pass→pass | 5,039 | 4,658 | -8% | 1 | 1 | 0% | 860 | 1,078 | +25% | 0 | 0 | — |
case-19 | pass→pass | 9,816 | 7,497 | -24% | 1 | 1 | 0% | 1,670 | 1,518 | -9% | 0 | 0 | — |
case-20 | pass→pass | 9,735 | 12,037 | +24% | 1 | 1 | 0% | 1,552 | 2,102 | +35% | 0 | 0 | — |
case-21 | pass→pass | 6,837 | 5,213 | -24% | 1 | 1 | 0% | 1,024 | 1,063 | +4% | 0 | 0 | — |
case-22 | pass→pass | 11,319 | 12,646 | +12% | 1 | 1 | 0% | 2,004 | 2,464 | +23% | 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. 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.