Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Find UI elements using natural language descriptions
.claude/skills/testdriverai-testdriver-locating-elements/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -60% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -47% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -33% | 0% |
<!-- Generated from locating-elements.mdx. DO NOT EDIT. -->
Use natural language to describe elements. Descriptions should be specific enough to locate the element, but not too-specific that they break with minor UI changes. For example:
javascriptawait testdriver.find('email input field'); await testdriver.find('first product card in the grid'); await testdriver.find('dropdown menu labeled "Country"');
<Info>TestDriver will cache found elements for improved performance on subsequent calls. Learn more about element caching here.</Info>
After finding an element, you can inspect its properties for debugging:
javascriptconst button = await testdriver.find('submit button'); console.log(button);
This outputs all element properties:
javascript{ description: 'submit button', found: true, x: 150, y: 300, coordinates: { x: 150, y: 300, centerX: 200, centerY: 320 }, threshold: 0.8, confidence: 0.95, similarity: 0.92, selector: 'button[type="submit"]', cache: { hit: true, strategy: 'pixel-diff', createdAt: '2025-01-15T10:30:00Z', diffPercent: 0.02, imageUrl: 'https://...' } }
Find and interact with multiple elements:
javascript// Find all matching elements const products = await testdriver.findAll('product card'); console.log(`Found ${products.length} products`); // Interact with each for (const product of products) { const title = await product.find('title text'); console.log('Product:', title.text); await product.find('add to cart button').click(); } // Or find specific element const firstProduct = products[0]; await firstProduct.click();
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 15,519 | 7,005 | -55% | 1 | 1 | 0% | 2,812 | 1,118 | -60% | 0 | 0 | — |
case-02 | fail→pass | 13,242 | 3,530 | -73% | 1 | 1 | 0% | 2,152 | 1,144 | -47% | 0 | 0 | — |
case-03 | fail→pass | 11,058 | 3,307 | -70% | 1 | 1 | 0% | 1,684 | 1,136 | -33% | 0 | 0 | — |
case-04 | fail→pass | 8,871 | 4,683 | -47% | 1 | 1 | 0% | 1,436 | 1,217 | -15% | 0 | 0 | — |
case-05 | fail→pass | 7,982 | 2,192 | -73% | 1 | 1 | 0% | 1,320 | 883 | -33% | 0 | 0 | — |
case-06 | fail→pass | 12,022 | 3,465 | -71% | 1 | 1 | 0% | 2,061 | 1,166 | -43% | 0 | 0 | — |
case-07 | fail→fail | 10,071 | 2,866 | -72% | 1 | 1 | 0% | 1,458 | 1,009 | -31% | 0 | 0 | — |
case-08 | pass→pass | 12,058 | 3,662 | -70% | 1 | 1 | 0% | 1,979 | 1,237 | -37% | 0 | 0 | — |
case-09 | pass→pass | 15,057 | 3,019 | -80% | 1 | 1 | 0% | 2,434 | 1,007 | -59% | 0 | 0 | — |
case-10 | fail→pass | 12,354 | 4,413 | -64% | 1 | 1 | 0% | 1,962 | 1,195 | -39% | 0 | 0 | — |
case-11 | pass→pass | 11,050 | 4,497 | -59% | 1 | 1 | 0% | 1,731 | 1,257 | -27% | 0 | 0 | — |
case-12 | pass→pass | 10,601 | 3,393 | -68% | 1 | 1 | 0% | 1,834 | 1,053 | -43% | 0 | 0 | — |
case-13 | pass→pass | 8,324 | 1,892 | -77% | 1 | 1 | 0% | 1,286 | 773 | -40% | 0 | 0 | — |
case-14 | pass→pass | 5,412 | 1,966 | -64% | 1 | 1 | 0% | 924 | 787 | -15% | 0 | 0 | — |
case-15 | pass→pass | 9,891 | 3,910 | -60% | 1 | 1 | 0% | 1,593 | 1,172 | -26% | 0 | 0 | — |
case-16 | pass→pass | 12,489 | 2,864 | -77% | 1 | 1 | 0% | 1,808 | 946 | -48% | 0 | 0 | — |
case-17 | pass→pass | 8,495 | 2,581 | -70% | 1 | 1 | 0% | 1,529 | 990 | -35% | 0 | 0 | — |
case-18 | fail→pass | 11,714 | 2,972 | -75% | 1 | 1 | 0% | 1,624 | 1,028 | -37% | 0 | 0 | — |
case-19 | pass→pass | 8,860 | 2,746 | -69% | 1 | 1 | 0% | 1,500 | 987 | -34% | 0 | 0 | — |
case-20 | pass→pass | 8,953 | 5,121 | -43% | 1 | 1 | 0% | 1,657 | 1,451 | -12% | 0 | 0 | — |
case-21 | pass→pass | 6,458 | 3,473 | -46% | 1 | 1 | 0% | 1,146 | 1,147 | +0% | 0 | 0 | — |
case-22 | pass→pass | 10,487 | 6,321 | -40% | 1 | 1 | 0% | 1,850 | 1,609 | -13% | 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.
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.