Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Decide whether to switch energy, broadband, or mobile providers — compare the real total cost, dodge the traps, and time it right. Use when asked should I switch energy/broadband/mobile providers, compare utility deals, is this a good energy tariff, or help me switch and save. Produces an apples-to-apples comparison (total annual cost, not headline rate), the traps to check (intro-then-jump pricing, exit fees, contract length), a switch/stay recommendation, the switching steps, and reminders to
.claude/skills/mohitagw15856-utility-switch-advisor/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 112% | 0% |
| case-23 | ✗→✓ | ▲ Improved | 64% | 0% |
| case-01 | ✓→✓ | = Same ✓ | -21% | 0% |
| case-02 | ✓→✓ | = Same ✓ | -17% | 0% |
Providers win by making comparison hard — teaser rates that jump, bundled discounts that expire, exit fees that trap you. This compares deals on what actually matters (total yearly cost for your usage), surfaces the traps, and gives a clear switch-or-stay call plus the steps to do it — while reminding you that live prices must be checked on a current comparison source.
Ask for these if not provided:
Total-cost compare (for your usage) | Option | ~Annual cost | Contract | Traps | |---|---|---|---| | Stay | x] | term] | price-rise?] | | Option] | x] | term] | intro-jump / exit fee] |
Recommendation: switch/stay] — save ~£/yr], watch catch]. Exit check: in-contract exit fee vs saving]. Switch steps: readings/details · timing · no-gap · keep number].
> Tariffs and deals change constantly — confirm today's actual prices on a live comparison source before committing.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 27,912 | 17,839 | -36% | 1 | 1 | 0% | 4,147 | 3,292 | -21% | 0 | 0 | — |
case-02 | pass→pass | 29,611 | 14,864 | -50% | 1 | 1 | 0% | 4,366 | 3,619 | -17% | 0 | 0 | — |
case-03 | pass→pass | 22,152 | 20,421 | -8% | 1 | 1 | 0% | 3,174 | 3,783 | +19% | 0 | 0 | — |
case-04 | pass→pass | 17,259 | 16,387 | -5% | 1 | 1 | 0% | 2,118 | 2,754 | +30% | 0 | 0 | — |
case-05 | pass→pass | 21,874 | 20,880 | -5% | 1 | 1 | 0% | 2,591 | 3,512 | +36% | 0 | 0 | — |
case-06 | pass→pass | 17,542 | 19,231 | +10% | 1 | 1 | 0% | 2,781 | 2,936 | +6% | 0 | 0 | — |
case-07 | pass→pass | 22,189 | 17,583 | -21% | 1 | 1 | 0% | 2,533 | 3,302 | +30% | 0 | 0 | — |
case-08 | pass→pass | 21,712 | 15,576 | -28% | 1 | 1 | 0% | 2,970 | 3,624 | +22% | 0 | 0 | — |
case-09 | pass→pass | 12,752 | 10,432 | -18% | 1 | 1 | 0% | 2,088 | 2,761 | +32% | 0 | 0 | — |
case-10 | fail→pass | 15,817 | 14,775 | -7% | 1 | 1 | 0% | 2,197 | 2,811 | +28% | 0 | 0 | — |
case-11 | fail→fail | 15,740 | 11,832 | -25% | 1 | 1 | 0% | 2,573 | 3,058 | +19% | 0 | 0 | — |
case-12 | fail→pass | 15,777 | 25,333 | +61% | 1 | 1 | 0% | 2,294 | 4,853 | +112% | 0 | 0 | — |
case-13 | pass→pass | 13,020 | 14,025 | +8% | 1 | 1 | 0% | 1,415 | 3,424 | +142% | 0 | 0 | — |
case-14 | fail→fail | 12,073 | 12,449 | +3% | 1 | 1 | 0% | 1,901 | 2,889 | +52% | 0 | 0 | — |
case-15 | pass→pass | 11,091 | 10,772 | -3% | 1 | 1 | 0% | 1,661 | 2,488 | +50% | 0 | 0 | — |
case-16 | pass→pass | 15,855 | 17,468 | +10% | 1 | 1 | 0% | 1,586 | 3,037 | +91% | 0 | 0 | — |
case-17 | pass→pass | 8,716 | 13,812 | +58% | 1 | 1 | 0% | 1,661 | 2,368 | +43% | 0 | 0 | — |
case-18 | pass→pass | 16,393 | 7,891 | -52% | 1 | 1 | 0% | 1,960 | 2,273 | +16% | 0 | 0 | — |
case-19 | pass→pass | 17,960 | 15,167 | -16% | 1 | 1 | 0% | 2,247 | 2,791 | +24% | 0 | 0 | — |
case-20 | pass→pass | 14,965 | 15,700 | +5% | 1 | 1 | 0% | 1,742 | 3,053 | +75% | 0 | 0 | — |
case-21 | pass→pass | 20,151 | 18,212 | -10% | 1 | 1 | 0% | 2,800 | 3,318 | +19% | 0 | 0 | — |
case-22 | pass→pass | 12,199 | 15,138 | +24% | 1 | 1 | 0% | 1,246 | 2,561 | +106% | 0 | 0 | — |
case-23 | fail→pass | 8,700 | 17,324 | +99% | 1 | 1 | 0% | 1,727 | 2,835 | +64% | 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. 23 cases were attempted. The headline lift of +13 percentage points is the difference between those two pass rates over the 23 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.