Observed arrival · 2026-09-24
Star Benchmark’s browser-agent obstacle course
Star Benchmark times browser agents as they work through signup, checkout, and ticket-sale tasks in Chrome.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Teams evaluating browser-control agents
- Worth noticing
- Runs are described as recording timed clicks and keystrokes, with a score and timeline available at /r/<run>.
Field notes
The listed tasks vary the kinds of friction an agent must handle: native controls on signup, custom widgets and a framed card form in checkout, then a queue and seat selection for tickets. The ticket task says Turnstile checks may appear on arrival, after seat selection, or before payment, while ticket insurance starts selected and must be declined.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue