Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when adding or testing flutter_rust_bridge feature flags, especially flags that require coverage for both enabled and disabled states or should have package-level config plus item-level #[frb(...)] overrides.
.claude/skills/fzyzcjy-frb-feature-flag/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -61% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -60% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -70% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -64% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -68% | 0% |
Use this skill when adding a feature flag, changing a feature flag, or writing coverage for a feature flag.
For feature flags requiring coverage for both enabled and disabled states:
flutter_rust_bridge.yaml.#[frb(...)] override when the behavior can reasonably be scoped to a Rust item.In frb_example/pure_dart, prefer tests that use the item-level #[frb(...)] override so enabled and disabled cases are both exercised in the same package.
Do not rely only on package-level config differences when both states can be covered clearly with item-level Rust APIs.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 13,656 | 5,354 | -61% | 1 | 1 | 0% | 2,580 | 1,164 | -55% | 0 | 0 | — |
case-02 | fail→pass | 13,234 | 4,032 | -70% | 1 | 1 | 0% | 2,338 | 903 | -61% | 0 | 0 | — |
case-03 | fail→pass | 18,177 | 6,188 | -66% | 1 | 1 | 0% | 3,323 | 1,337 | -60% | 0 | 0 | — |
case-04 | pass→pass | 11,335 | 5,170 | -54% | 1 | 1 | 0% | 1,999 | 1,070 | -46% | 0 | 0 | — |
case-05 | pass→pass | 10,344 | 6,132 | -41% | 1 | 1 | 0% | 2,107 | 1,330 | -37% | 0 | 0 | — |
case-14 | fail→pass | 10,061 | 2,117 | -79% | 1 | 1 | 0% | 1,794 | 540 | -70% | 0 | 0 | — |
case-06 | pass→pass | 8,784 | 4,979 | -43% | 1 | 1 | 0% | 1,708 | 1,027 | -40% | 0 | 0 | — |
case-07 | fail→pass | 12,236 | 3,511 | -71% | 1 | 1 | 0% | 2,307 | 835 | -64% | 0 | 0 | — |
case-08 | pass→pass | 11,508 | 3,991 | -65% | 1 | 1 | 0% | 1,644 | 676 | -59% | 0 | 0 | — |
case-09 | fail→pass | 13,199 | 2,697 | -80% | 1 | 1 | 0% | 2,038 | 658 | -68% | 0 | 0 | — |
case-10 | pass→pass | 10,753 | 2,652 | -75% | 1 | 1 | 0% | 2,039 | 621 | -70% | 0 | 0 | — |
case-11 | fail→pass | 14,047 | 3,278 | -77% | 1 | 1 | 0% | 2,590 | 776 | -70% | 0 | 0 | — |
case-12 | pass→pass | 12,617 | 2,004 | -84% | 1 | 1 | 0% | 2,108 | 530 | -75% | 0 | 0 | — |
case-13 | pass→pass | 14,143 | 3,835 | -73% | 1 | 1 | 0% | 2,371 | 768 | -68% | 0 | 0 | — |
case-15 | fail→pass | 15,638 | 3,715 | -76% | 1 | 1 | 0% | 2,950 | 880 | -70% | 0 | 0 | — |
case-16 | pass→pass | 14,049 | 1,912 | -86% | 1 | 1 | 0% | 2,568 | 527 | -79% | 0 | 0 | — |
case-17 | fail→pass | 9,543 | 3,727 | -61% | 1 | 1 | 0% | 1,614 | 740 | -54% | 0 | 0 | — |
case-18 | fail→pass | 8,417 | 1,734 | -79% | 1 | 1 | 0% | 1,585 | 457 | -71% | 0 | 0 | — |
case-19 | pass→pass | 12,067 | 2,382 | -80% | 1 | 1 | 0% | 2,088 | 598 | -71% | 0 | 0 | — |
case-20 | fail→pass | 6,863 | 1,379 | -80% | 1 | 1 | 0% | 1,193 | 333 | -72% | 0 | 0 | — |
case-21 | pass→pass | 8,457 | 1,439 | -83% | 1 | 1 | 0% | 1,653 | 379 | -77% | 0 | 0 | — |
case-22 | fail→pass | 18,102 | 3,727 | -79% | 1 | 1 | 0% | 3,451 | 910 | -74% | 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 +50 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.