A browser automation agent that heals its own path

Script-based browser automation anchors every action to a brittle selector or XPath, so a renamed label or a new CSS class breaks the run and a human fixes the script. A browser automation agent anchors to intent, the user-facing element like "the Submit button," and re-resolves it when the page shifts. It waits on the rendered viewport instead of a network or DOM signal, works through unexpected modals and redirects up to 50 steps, and returns a verified pass or fail. You write what you want, not the path to get there.

Yes. Autoheal is the core of the agent. Because each action is anchored to the user-facing element rather than a fixed selector, a reworded label, a new class, or a shifted layout does not stall the flow. The agent re-resolves the element and keeps going, and it pushes through cookie banners, modals, and redirects on the way. Cosmetic drift that would break a Selenium or Playwright script is handled without a human touching the automation, while the pass condition it must verify stays exactly the same.

Yes, and the deterministic verdict is what makes a non-deterministic model safe to gate on. Authenticate with your TestMu AI credentials, pass --headless, --agent, and --timeout, then key the pipeline off the evidence-backed exit code: 0 when the expected state is verified, 1 on a failed assertion, 2 on an error such as auth, and 3 on timeout. The same binary runs headed on your laptop and headless on GitHub Actions, GitLab CI, Jenkins, or Bitbucket, so a broken journey stops the release.

Yes. Because the verdict is machine-readable, AI coding agents can drive the browser automation agent and close their own loop. Run Kane CLI with the --agent flag and it streams NDJSON, one typed event per line, ending in a run_end event carrying the verdict, an evidence summary, extracted values, and a dashboard link. Claude Code, Cursor, Codex, and Gemini CLI parse that stream to check the UI they generated. Point your agent at the guide at testmuai.com/kane-cli/agents.md and it installs, authenticates, and drives the flow on its own.

No lock-in. Any completed run exports to native Python Playwright with --code-export --code-language python. The exported code uses the testmu library, which keeps the original intent-driven step alongside the generated code, so the suite keeps the agent's autoheal resilience even when run as native Playwright. You can also import existing Playwright or Selenium scripts, so you bring your suite in and take it out whenever you want.

When one Chrome on your laptop is not enough, the same objective runs on Browser Cloud, the browser infrastructure built for AI agents. It provisions real, full-featured Chrome sessions on demand and runs many in parallel, with a built-in tunnel to reach localhost and staging, automatic session video and logs, and persistent login state, all on the cloud that powers 1.5 billion tests a year for 18,000+ enterprises. Kane CLI is free to install, with 100 starter credits and no credit card, and cloud runs are billed against your plan only when you scale.