RunAgent SuperBrowser: A Theory of Autonomous Web Navigation Grounded in Human Browsing Behaviour
2026-06-08 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors created SUPERBROWSER, a web agent designed to browse websites like a human does by focusing only on important parts rather than everything on the screen. They broke down the process into seeing clickable areas first, making plans and decisions through separate roles, and remembering just key information like goals and recent actions. Their system also manages memory by regularly clearing out old or unneeded data. Tested on a set of web tasks, SUPERBROWSER performed very well, outperforming other research tools. The authors suggest the success comes from applying human-like browsing principles consistently across the system.
web navigation agentlanguage modelbounding-box detectionstrategic reasoningoperational actionsmemory managementChrome DevTools ProtocolPuppeteercognitive architectureMind2Web benchmark
Authors
Radeen Mostafa, Sawradip Saha
Abstract
We present SUPERBROWSER, an autonomous web-navigation agent designed against a single guiding hypothesis: a web agent should browse the way a person browses. A human reading a page does not retain every pixel they have seen; they look at a few candidate targets, decide on one, and remember only what is needed to keep the goal alive. We operationalize this perception-cognition-action triad as three coupled mechanisms. First, a vision-first bounding-box pipeline labels candidate interactive regions on every screenshot and feeds them, asynchronously prefetched, to the language model so that the "eye" precedes the "hand". Second, a three-role brain -- an Orchestrator that classifies and routes, a Planner that evaluates progress every few steps, and a Worker that emits per-step actions -- separates strategic from operational reasoning. Third, a structured Ledger stores only what a person would: the goal, the last three actions, a small set of facts and dead-ends, and a handful of checkpoints; a six-phase eviction loop systematically discards stale screenshots, state blobs, and reasoning traces from the live context. Action execution is a three-tier click cascade (Chrome DevTools Protocol to Puppeteer to scripted) with humanized Bezier motion, plus a chevron-aware bounding-box snapper that resolves the "small arrow beside a large label" ambiguity. On the Mind2Web Hard benchmark (66 tasks), SUPERBROWSER attains 89.47% success, placing third overall and ahead of every published open/research browser-agent baseline by a large margin. We argue that the gain comes not from any single trick but from the consistent application of a cognitive contract throughout the system.