Crossbrowsertesting.com Screenshot Comparisons API

{
  "service_type": "platform",
  "base_url": "https://crossbrowsertesting.com",
  "auth_method": "api_key",
  "auth_config": {
    "notes": "CrossBrowserTesting historically uses HTTP Basic auth with a username (email) and auth key, or an API key passed as a query/header parameter. Exact scheme should be confirmed against current SmartBear documentation."
  },
  "endpoints": [],
  "pricing_model": {
    "type": "subscription",
    "details": {
      "notes": "CrossBrowserTesting is a commercial platform (now part of SmartBear) sold via subscription plans based on parallel sessions, minutes, and features. Screenshot comparison is a feature of the platform rather than a separately priced API."
    }
  },
  "rate_limits": {},
  "capabilities": [
    "Cross-browser screenshot capture across browser/OS/device combinations",
    "Visual comparison of two screenshots to detect layout differences",
    "Comparison of a full multi-browser screenshot test against a single baseline browser",
    "Visual regression testing / layout diffing",
    "Cloud-based browser and device test grid",
    "Selenium / automated testing integration",
    "Manual and automated browser testing"
  ],
  "raw_analysis": "CrossBrowserTesting.com is a cloud-based cross-browser and cross-device testing platform, acquired by SmartBear and now offered under the SmartBear testing portfolio. Its core offering is a cloud grid of real browsers, OSes, and mobile devices for manual and automated (Selenium/Appium) testing, plus visual regression capabilities.\n\nThe 'Screenshot Comparisons API' described here provides programmatic visual diffing: (1) compare two screenshots for layout differences, (2) compare a full multi-browser screenshot test against a single baseline browser, and (3) additional comparison modes (description truncated). This is effectively a visual regression testing API layered on top of their screenshot-capture service.\n\nMaturity: CrossBrowserTesting is a well-established, mature commercial platform with years of operation prior to the SmartBear acquisition; the API is a documented feature of the product. It is aimed at QA engineers, front-end developers, and teams doing cross-browser and visual regression testing.\n\nAuth: The platform traditionally uses an API key / auth key associated with an account, typically combined with the account email via HTTP Basic auth. Exact current scheme should be verified against SmartBear API docs since the product has been folded into SmartBear's documentation.\n\nIntegrations: Works with Selenium, CI/CD pipelines, and their own screenshot/comparison tooling. Pricing is subscription-based with tiers gated by parallel sessions and usage volume; there is no free public API tier per se, though trials are typically available.\n\nNote: The provided description is truncated, so the exact endpoint list and full parameter schema could not be extracted. Endpoints section is left empty pending access to the full API reference."
}

0/3 tests passed

```json
{
  "overall": 38,
  "dimensions": {
    "token_efficiency": 6.0,
    "first_try_success": 3.0,
    "response_parseability": 5.0,
    "error_clarity": 4.0,
    "doc_quality": 4.5,
    "auth_simplicity": 3.0,
    "latency": 2.0,
    "consistency": 4.0
  },
  "pricing_normalized": {
    "model": "subscription",
    "notes": "Sales-led enterprise subscription (SmartBear). No public self-serve price list captured; parallel sessions / minutes tiers. Not agent-friendly pricing — requires human sales contact."
  },
  "issues": [
    "DNS resolution failed for hostname across all three checks — the domain could not be reached at test time (possibly retired/rebranded to SmartBear visual testing).",
    "No public API-first pricing; subscription is quota-based on parallel sessions/minutes, which is hard for agents to estimate cost.",
    "Onboarding is sales/docs-gated rather than self-serve sign-up — poor first-try success for agents.",
    "No llms.txt and no robots.txt retrieved — no machine-readable or agent-guidance surface exposed.",
    "Screenshot comparison is a bundled feature, not a clean standalone API — agents can't easily reason about it as a discrete tool."
  ],
  "recommendations": [
    "Do not recommend until domain resolves and a live, reachable endpoint/doc surface is confirmed.",
    "Prefer modern, API-first visual regression tools (e.g., Percy, Chromatic, Playwright/BackstopJS local) for agent use cases — self-serve, clear pricing, structured APIs.",
    "If CrossBrowserTesting is needed specifically, direct users to the SmartBear product page and verify current domain/redirect before relying on it.",
    "Request a public machine-readable pricing/API spec and status page to raise first_try_success and consistency scores."
  ]
}
```