Benchmarking Browsers and Web Tools for AI Agents. An evaluation harness for Web Fetch / Search / Scraping / Crawling, and Browser Trajectory Replay.
evaluations web-scraping serp web-crawling browser-automation ai-agents moli headless-browser browser-benchmark playwright webarena computer-use browser-use agent-evaluation web-fetch trajectory-replay webmainbench
-
Updated
Nov 4, 2026 - Python