Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Write an implementation plan that leads with the decisions the user is most likely to change, and buries the mechanical work at the bottom. Use when planning is requested before a build, especially after brainstorming or an interview.
.claude/skills/neeeophytee-implementation-plan/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -9% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -26% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -15% | 0% |
A plan's job is not to prove you thought of everything. It's to put the reversible-but-expensive decisions in front of the user while changing them is still free. Order the plan by likelihood-of-tweaking, not by build order.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 26,268 | 25,246 | -4% | 1 | 1 | 0% | 4,123 | 2,959 | -28% | 0 | 0 | — |
case-02 | fail→pass | 21,953 | 17,419 | -21% | 1 | 1 | 0% | 3,246 | 2,951 | -9% | 0 | 0 | — |
case-03 | fail→pass | 25,500 | 17,819 | -30% | 1 | 1 | 0% | 3,978 | 2,936 | -26% | 0 | 0 | — |
case-04 | pass→pass | 20,250 | 25,261 | +25% | 1 | 1 | 0% | 3,227 | 4,295 | +33% | 0 | 0 | — |
case-05 | pass→fail | 14,902 | 15,672 | +5% | 1 | 1 | 0% | 2,897 | 3,166 | +9% | 0 | 0 | — |
case-06 | pass→pass | 10,724 | 12,265 | +14% | 1 | 1 | 0% | 1,622 | 2,134 | +32% | 0 | 0 | — |
case-07 | pass→pass | 14,766 | 14,944 | +1% | 1 | 1 | 0% | 3,315 | 3,549 | +7% | 0 | 0 | — |
case-08 | fail→pass | 21,814 | 18,646 | -15% | 1 | 1 | 0% | 3,516 | 3,134 | -11% | 0 | 0 | — |
case-09 | fail→fail | 16,218 | 18,540 | +14% | 1 | 1 | 0% | 2,625 | 3,120 | +19% | 0 | 0 | — |
case-10 | fail→pass | 22,960 | 19,615 | -15% | 1 | 1 | 0% | 3,913 | 3,309 | -15% | 0 | 0 | — |
case-11 | fail→pass | 20,731 | 17,622 | -15% | 1 | 1 | 0% | 3,169 | 3,082 | -3% | 0 | 0 | — |
case-12 | pass→pass | 27,533 | 25,000 | -9% | 1 | 1 | 0% | 4,974 | 5,168 | +4% | 0 | 0 | — |
case-13 | fail→pass | 20,877 | 15,780 | -24% | 1 | 1 | 0% | 3,363 | 2,702 | -20% | 0 | 0 | — |
case-14 | fail→fail | 33,910 | 19,738 | -42% | 1 | 1 | 0% | 6,185 | 3,307 | -47% | 0 | 0 | — |
case-15 | pass→pass | 10,987 | 15,464 | +41% | 1 | 1 | 0% | 1,745 | 2,452 | +41% | 0 | 0 | — |
case-16 | fail→fail | 20,990 | 15,620 | -26% | 1 | 1 | 0% | 3,369 | 2,691 | -20% | 0 | 0 | — |
case-17 | fail→pass | 19,058 | 12,514 | -34% | 1 | 1 | 0% | 3,291 | 2,253 | -32% | 0 | 0 | — |
case-18 | fail→fail | 19,320 | 35,544 | +84% | 1 | 1 | 0% | 3,043 | 2,443 | -20% | 0 | 0 | — |
case-19 | fail→fail | 40,885 | 17,093 | -58% | 1 | 1 | 0% | 3,898 | 3,113 | -20% | 0 | 0 | — |
case-20 | fail→pass | 20,363 | 13,071 | -36% | 1 | 1 | 0% | 3,268 | 2,355 | -28% | 0 | 0 | — |
case-21 | fail→pass | 22,667 | 12,670 | -44% | 1 | 1 | 0% | 3,756 | 2,321 | -38% | 0 | 0 | — |
case-22 | fail→pass | 20,355 | 17,021 | -16% | 1 | 1 | 0% | 3,371 | 3,025 | -10% | 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 +45 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.