Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Set up Spectron (deprecated) tests for legacy Electron application testing
.claude/skills/a5c-ai-spectron-test-setup/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -62% | 0% |
| case-02 | ✓→✗ | ▼ Worse | -30% | 0% |
| case-01 | ✓→✓ | = Same ✓ | -17% | 0% |
| case-03 | ✓→✓ | = Same ✓ | -16% | 0% |
Set up Spectron tests for legacy Electron applications. Note: Spectron is deprecated; use Playwright for new projects.
json{ "type": "object", "properties": { "projectPath": { "type": "string" }, "electronPath": { "type": "string" } }, "required": ["projectPath"] }
javascriptconst Application = require('spectron').Application; const path = require('path'); describe('Application launch', function () { beforeEach(function () { this.app = new Application({ path: require('electron'), args: [path.join(__dirname, '..')] }); return this.app.start(); }); afterEach(function () { if (this.app && this.app.isRunning()) { return this.app.stop(); } }); it('shows initial window', function () { return this.app.client.getWindowCount().then(count => { expect(count).toBe(1); }); }); });
Consider migrating to playwright-electron-config for modern Electron testing.
playwright-electron-configelectron-mock-factory| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 9,738 | 5,957 | -39% | 1 | 1 | 0% | 1,781 | 1,478 | -17% | 0 | 0 | — |
case-02 | pass→fail | 12,076 | 7,601 | -37% | 1 | 1 | 0% | 2,256 | 1,578 | -30% | 0 | 0 | — |
case-03 | pass→pass | 9,755 | 6,172 | -37% | 1 | 1 | 0% | 1,866 | 1,568 | -16% | 0 | 0 | — |
case-04 | pass→pass | 7,905 | 6,012 | -24% | 1 | 1 | 0% | 1,541 | 1,414 | -8% | 0 | 0 | — |
case-05 | fail→pass | 5,360 | 1,889 | -65% | 1 | 1 | 0% | 917 | 623 | -32% | 0 | 0 | — |
case-06 | pass→pass | 8,472 | 1,943 | -77% | 1 | 1 | 0% | 1,350 | 694 | -49% | 0 | 0 | — |
case-07 | pass→pass | 14,634 | 8,258 | -44% | 1 | 1 | 0% | 2,432 | 1,933 | -21% | 0 | 0 | — |
case-08 | pass→pass | 11,970 | 9,774 | -18% | 1 | 1 | 0% | 2,441 | 2,511 | +3% | 0 | 0 | — |
case-09 | pass→pass | 8,967 | 4,229 | -53% | 1 | 1 | 0% | 1,468 | 1,162 | -21% | 0 | 0 | — |
case-10 | pass→pass | 8,548 | 5,730 | -33% | 1 | 1 | 0% | 1,643 | 1,291 | -21% | 0 | 0 | — |
case-11 | pass→pass | 11,382 | 6,690 | -41% | 1 | 1 | 0% | 2,186 | 1,627 | -26% | 0 | 0 | — |
case-12 | pass→pass | 8,681 | 4,646 | -46% | 1 | 1 | 0% | 1,656 | 1,253 | -24% | 0 | 0 | — |
case-13 | pass→pass | 7,784 | 3,980 | -49% | 1 | 1 | 0% | 1,461 | 1,127 | -23% | 0 | 0 | — |
case-14 | pass→pass | 5,787 | 6,209 | +7% | 1 | 1 | 0% | 1,163 | 1,532 | +32% | 0 | 0 | — |
case-15 | pass→pass | 8,856 | 4,291 | -52% | 1 | 1 | 0% | 1,675 | 1,192 | -29% | 0 | 0 | — |
case-16 | pass→pass | 13,548 | 6,277 | -54% | 1 | 1 | 0% | 2,342 | 1,743 | -26% | 0 | 0 | — |
case-17 | pass→pass | 8,847 | 4,127 | -53% | 1 | 1 | 0% | 1,467 | 1,102 | -25% | 0 | 0 | — |
case-18 | fail→pass | 11,361 | 2,548 | -78% | 1 | 1 | 0% | 1,921 | 728 | -62% | 0 | 0 | — |
case-19 | pass→pass | 9,691 | 1,678 | -83% | 1 | 1 | 0% | 1,771 | 621 | -65% | 0 | 0 | — |
case-20 | pass→pass | 10,728 | 8,148 | -24% | 1 | 1 | 0% | 2,056 | 2,129 | +4% | 0 | 0 | — |
case-21 | pass→pass | 13,353 | 7,722 | -42% | 1 | 1 | 0% | 2,742 | 1,902 | -31% | 0 | 0 | — |
case-22 | pass→pass | 10,369 | 8,204 | -21% | 1 | 1 | 0% | 2,183 | 1,727 | -21% | 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 +5 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.