testing-browser-compatibilityClaude Skill
Test across multiple browsers and devices for cross-browser compatibility.
| name | testing-browser-compatibility |
| description | Test across multiple browsers and devices for cross-browser compatibility. Use when ensuring cross-browser or device compatibility with BrowserStack, Sauce Labs, LambdaTest, or Kobiton. Trigger with phrases like "test browser compatibility", "check cross-browser", "validate on browsers", "test on real devices", "kobiton test". |
| allowed-tools | Read, Write, Edit, Grep, Glob, Bash(npx playwright:*), Bash(npm:*), Bash(curl:*) |
| version | 2.0.0 |
| author | Jeremy Longshore <jeremy@intentsolutions.io> |
| license | MIT |
| compatible-with | claude-code, codex, openclaw |
| tags | ["testing","cross-browser","mobile-testing","real-device","cloud-testing"] |
Browser Compatibility Tester
Table of Contents
Overview | Instructions (Local / Cloud) | Examples | Error Handling | Output | Resources
Overview
Test web applications across multiple browsers, rendering engines, and real devices. Validates CSS rendering, JavaScript API support, layout consistency, and interactive behavior across Chromium (Chrome, Edge), Gecko (Firefox), and WebKit (Safari) -- locally with Playwright or on real devices via BrowserStack, Sauce Labs, LambdaTest, or Kobiton.
Prerequisites
- Playwright installed (
npx playwright install --with-deps) and application running at a test URL - For cloud testing: provider credentials in environment variables (see
${CLAUDE_SKILL_DIR}/references/cloud-providers.md)
Instructions
Mode 1: Local Testing (Playwright)
Default mode. Zero cloud accounts needed.
-
Define the browser matrix from project
browserslistconfig or use defaults:- Desktop: Chrome (latest), Firefox (latest), Safari (latest), Edge (latest)
- Mobile: iPhone 14 (WebKit), Pixel 7 (Chromium)
- Viewports: 375px, 768px, 1280px, 1920px
-
Scan the codebase for compatibility risks:
- Grep for modern JS APIs (
IntersectionObserver,structuredClone,Array.at(),Promise.withResolvers()) - Grep for modern CSS (
container queries,has(),@layer,subgrid,color-mix()) - Cross-reference against caniuse data; flag usage without polyfills or
@supports
- Grep for modern JS APIs (
-
Write compatibility-focused tests:
- Layout: key elements render at expected positions/sizes per viewport
- CSS features: modern features degrade gracefully behind
@supports - JS APIs: polyfills load in older browsers; form inputs (date, color, range) across engines
- Accessibility: run axe-core per browser (
@axe-core/playwright)
-
Execute and capture results:
npx playwright test --project=chromium --project=firefox --project=webkit- Screenshots per browser for visual comparison
- Video traces for failing tests
Mode 2: Cloud Real-Device Testing
Applies when real physical devices, broader OS coverage, or carrier network conditions are required beyond what Playwright emulation can replicate. Read ${CLAUDE_SKILL_DIR}/references/cloud-providers.md for full auth, API, and capabilities details.
Provider selection:
| Need | Provider |
|---|---|
| Broadest browser/OS matrix (3,000+ combos) | BrowserStack |
| Enterprise CI/CD, Sauce Connect tunnel | Sauce Labs |
| Auto-healing selectors, smart testing | LambdaTest |
| Real physical devices, scriptless automation | Kobiton |
Never hardcode credentials. Set provider env vars (BROWSERSTACK_USERNAME/ACCESS_KEY, SAUCE_USERNAME/ACCESS_KEY, LT_USERNAME/ACCESS_KEY, KOBITON_USERNAME/API_KEY).
- Verify credentials are set for the chosen provider
- Query available devices/browsers via provider API
- Configure WebDriver or Appium capabilities (see
${CLAUDE_SKILL_DIR}/references/cloud-providers.md) - Execute tests against cloud grid
- Retrieve session artifacts (screenshots, video, logs, network HAR)
- Aggregate results into compatibility report (CI/CD patterns:
${CLAUDE_SKILL_DIR}/references/ci-cd-integration.md)
Browser-Specific Checks
- Safari: date input formatting, scroll behavior, backdrop-filter, PWA manifest, position: sticky in overflow
- Firefox: scrollbar styling, gap in flexbox, subpixel rendering, print stylesheets
- Mobile: touch events, viewport meta, safe area insets, virtual keyboard resize
Pre-built device matrices: ${CLAUDE_SKILL_DIR}/references/device-matrix.md (top 10, mobile-first, enterprise, Kobiton real-device).
Examples
Playwright multi-browser config:
import { defineConfig, devices } from '@playwright/test'; export default defineConfig({ projects: [ { name: 'chromium', use: { ...devices['Desktop Chrome'] } }, { name: 'firefox', use: { ...devices['Desktop Firefox'] } }, { name: 'webkit', use: { ...devices['Desktop Safari'] } }, { name: 'mobile-chrome', use: { ...devices['Pixel 7'] } }, { name: 'mobile-safari', use: { ...devices['iPhone 14'] } }, ], });
Cross-browser layout test:
test('nav renders correctly across browsers', async ({ page }) => { await page.goto('/'); const nav = page.locator('nav'); await expect(nav).toBeVisible(); const box = await nav.boundingBox(); expect(box.width).toBeGreaterThan(300); });
Kobiton real-device capabilities:
{ "platformName": "iOS", "appium:deviceName": "iPhone 15 Pro", "appium:platformVersion": "17", "browserName": "Safari", "kobiton:options": { "sessionName": "Safari Compat Test", "deviceGroup": "KOBITON", "captureScreenshots": true } }
Error Handling
| Error | Cause | Solution |
|---|---|---|
| WebKit fails, Chromium passes | CSS property unsupported in Safari | Add -webkit- prefix or @supports fallback |
| Date input renders differently | Browsers implement <input type="date"> differently | Use custom date picker component |
| Test passes locally, fails on cloud | Real device rendering differs from emulation | Run critical paths on real devices for final validation |
| Kobiton device unavailable | Device in use or offline | Query GET /v1/devices for online devices; use deviceGroup for flexible matching |
| Cloud session timeout | Long test on slow device | Increase sessionTimeout; split into smaller test files |
Output
- Playwright config with multi-browser projects and test files in
tests/compatibility/ - Compatibility matrix report (pass/fail per browser, viewport, device)
- Screenshots per browser for visual diff; unsupported API list with polyfill recommendations
- Cloud session URLs with video replay links (when using cloud providers)
Resources
- Playwright: https://playwright.dev/docs/browsers | Can I Use: https://caniuse.com/
- Cloud providers: https://www.browserstack.com/automate | https://docs.saucelabs.com/ | https://www.lambdatest.com/support/docs/ | https://api.kobiton.com/docs/
- MDN Compat Data: https://github.com/mdn/browser-compat-data
Similar Claude Skills & Agent Workflows
end-to-end-tests
after making changes, run end-to-end tests to ensure that the product still works
test-coverage-improver
Improve test coverage in the OpenAI Agents Python repository: run `make coverage`, inspect coverage artifacts, identify low-coverage files, propose high-impact tests, and confirm with the user before writing tests.
code-change-verification
Run the mandatory verification stack when changes affect runtime code, tests, or build/test behavior in the OpenAI Agents Python repository.
testing-python
Write and evaluate effective Python tests using pytest.
testing
Run and troubleshoot tests for DBHub, including unit tests, integration tests with Testcontainers, and database-specific tests.
n8n-validation-expert
Interpret validation errors and guide fixing them.