Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Build a veterinary treatment plan with tiered options and a cost estimate to discuss with a pet owner. Use when asked to prepare a treatment plan, create an estimate for an owner, present diagnostic/treatment options, or have the cost conversation in a vet practice. Produces a clear plan (recommended vs. acceptable-alternative vs. minimum), line-item cost ranges, the medical rationale in plain language, and how to frame the money conversation with empathy so the owner can make an informed, unpre
.claude/skills/mohitagw15856-treatment-plan-estimate/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 46% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 52% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 26% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 37% | 0% |
The hardest conversation in a vet practice is money, and avoiding it hurts everyone — the owner feels ambushed by the bill, the pet gets under-treated, the practice eats the loss. This skill builds a plan with honest options at different price points and frames the estimate so the owner can choose with full information and no shame.
Given the presentation and the recommended workup/treatment, produce the full plan and estimate — offer tiered options, use plain owner-facing language, and label cost figures as estimate ranges to be confirmed by the practice. Never diagnose beyond the information given or state prices as exact.
Ask for (if not provided, else infer and label):
What's going on (or what we need to find out), and why each step is recommended — jargon translated for the owner.
| Tier | What it includes | Why | Estimated cost (range) | |---|---|---|---| | Recommended (gold standard) | full workup/treatment | best outcome/certainty | | | Acceptable alternative | the pragmatic middle | good care within constraints | | | Minimum / palliative | comfort + essentials | when budget is tight or prognosis guarded | |
Line items with ranges; note what could change the total (findings, complications).
How to present it with empathy: normalize asking about cost, present options without judgment, avoid pressure, and offer any payment options/financing the practice supports. Make clear a "no" to the top tier is a valid, respected choice.
What the owner is agreeing to, the deposit/authorization, and the decision point (what happens if findings change the plan mid-procedure).
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 23,100 | 26,048 | +13% | 1 | 1 | 0% | 2,445 | 3,576 | +46% | 0 | 0 | — |
case-02 | fail→pass | 18,574 | 22,320 | +20% | 1 | 1 | 0% | 2,237 | 3,410 | +52% | 0 | 0 | — |
case-03 | fail→pass | 24,727 | 24,199 | -2% | 1 | 1 | 0% | 2,921 | 3,438 | +18% | 0 | 0 | — |
case-04 | pass→pass | 22,221 | 25,392 | +14% | 1 | 1 | 0% | 2,329 | 3,386 | +45% | 0 | 0 | — |
case-05 | fail→pass | 23,181 | 26,314 | +14% | 1 | 1 | 0% | 2,711 | 3,426 | +26% | 0 | 0 | — |
case-06 | fail→pass | 21,624 | 28,036 | +30% | 1 | 1 | 0% | 3,019 | 4,121 | +37% | 0 | 0 | — |
case-07 | fail→pass | 25,730 | 22,174 | -14% | 1 | 1 | 0% | 2,728 | 3,476 | +27% | 0 | 0 | — |
case-08 | pass→pass | 20,185 | 20,033 | -1% | 1 | 1 | 0% | 2,289 | 2,722 | +19% | 0 | 0 | — |
case-09 | fail→pass | 34,308 | 27,690 | -19% | 1 | 1 | 0% | 2,536 | 4,133 | +63% | 0 | 0 | — |
case-10 | pass→pass | 23,639 | 23,645 | +0% | 1 | 1 | 0% | 2,448 | 3,558 | +45% | 0 | 0 | — |
case-11 | pass→pass | 19,006 | 25,593 | +35% | 1 | 1 | 0% | 1,984 | 3,277 | +65% | 0 | 0 | — |
case-12 | fail→pass | 34,199 | 18,774 | -45% | 1 | 1 | 0% | 3,192 | 3,590 | +12% | 0 | 0 | — |
case-13 | pass→pass | 24,298 | 24,913 | +3% | 1 | 1 | 0% | 3,300 | 3,875 | +17% | 0 | 0 | — |
case-14 | pass→pass | 20,023 | 22,986 | +15% | 1 | 1 | 0% | 2,363 | 4,015 | +70% | 0 | 0 | — |
case-15 | pass→pass | 21,828 | 22,992 | +5% | 1 | 1 | 0% | 2,717 | 3,406 | +25% | 0 | 0 | — |
case-16 | fail→pass | 25,587 | 21,287 | -17% | 1 | 1 | 0% | 2,580 | 3,052 | +18% | 0 | 0 | — |
case-17 | fail→pass | 19,342 | 28,653 | +48% | 1 | 1 | 0% | 2,249 | 3,639 | +62% | 0 | 0 | — |
case-18 | fail→pass | 28,249 | 24,279 | -14% | 1 | 1 | 0% | 3,244 | 3,651 | +13% | 0 | 0 | — |
case-19 | pass→pass | 23,607 | 23,608 | +0% | 1 | 1 | 0% | 3,286 | 3,610 | +10% | 0 | 0 | — |
case-20 | pass→fail | 34,285 | 23,188 | -32% | 1 | 1 | 0% | 3,316 | 3,329 | +0% | 0 | 0 | — |
case-21 | pass→fail | 19,425 | 28,797 | +48% | 1 | 1 | 0% | 2,285 | 4,435 | +94% | 0 | 0 | — |
case-22 | pass→fail | 18,068 | 33,394 | +85% | 1 | 1 | 0% | 2,876 | 4,468 | +55% | 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 +36 percentage points is the difference between those two pass rates over the 22 comparable cases. 3 cases got worse with the skill loaded, and they are 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.