Stop shipping on vibes.
Paste a URL and one sentence. The AI planner explains what it will test and why. A real browser runs it. A vision model explains whether the screenshots prove the goal. You walk away with a shareable proof page — and a release-health entry that tracks the next deploy.
AI plan · browser evidence · vision verdict · shareable proof
- 01 / PLAN
AI plans the test.
Paste a URL and one sentence. The planner returns a structured plan with assumptions, missing context, browser steps, and a self-rated confidence — before any browser spins up.
- 02 / RUN
A real browser uses it.
Playwright MCP executes the plan in Chromium, Firefox, or WebKit. Every step is asserted; screenshots, console, and network evidence are captured per step.
- 03 / EVIDENCE
Vision reads the proof.
A vision model reviews the screenshots, returns a verdict with confidence and limitations, and adds the run to your release-health history. Share a token-gated proof page anyone can open.
Start with three runs.
Free stays free. When you need more, Starter and Growth include a one-time 7-day trial. Release health, bug history, and trend analysis unlock on Growth and Scale.
Compare plans →Annual billing saves 10%| Free | $0/mo | 3 runs each month · card required · no charge |
|---|---|---|
| Starter | $79/mo | For one builder shipping regularly |
| Growth | $249/mo | Multi-browser, AI credits, basic release health |
| Scale | $799/mo | High-volume runs, full release health, bug history |