Detection

Every automation system — defined, scored, explained

We don’t ship a mystery “risk number.” Each signal maps to a founder category and a recommendation you can act on. Free tier includes the full suite.

Founder categories

What your console and API detection pack show.

  • Not a real browser

    Automation, headless Chrome, or anti-detect tools.

  • AI agent

    AI browser agents or agent-style extensions.

  • Hiding their location

    Datacenter IP, VPN, or timezone/language mismatches.

  • Throwaway or risky email

    Disposable inbox or identity-graph exposure links.

  • Repeat offender

    Linked to prior abuse or signup flood velocity.

Automation detectors

Why each family exists — so you know what “not a real browser” actually means.

  • Headless / webdriver

    navigator.webdriver, HeadlessChrome UA, missing window.chrome — classic bot runtimes.

  • Playwright / Puppeteer / CDP

    Framework leftovers on window/document, exposeFunction hooks, ChromeDriver CDP keys.

  • Anti-detect browsers

    Prototype lies, UA/platform mismatches, Client Hints conflicts — stealth kits that fail consistency checks.

  • Behavioral

    No interaction, synthetic events, teleport mouse, linear paths, robotic typing — humans don’t move like scripts.

  • Cloudflare Bot Management

    Verified bots, low bot scores, detection IDs, JA3 — server-side when the edge sees the request.

Recommendations in the product

Every check returns an action (allow / challenge / shadow / block / review), a plain-English title, and bullets from visit drift + top signals. Your dashboard and review queue surface them so operators aren’t guessing.