2607.02850 ClawBench: An Evidence Brief on Everyday Online Web-Agent Evaluation
This evidence brief summarizes the publicly documented ClawBench benchmark for evaluating browser agents on everyday online workflows. ClawBench evaluates 153 tasks across 144 production websites and 15 life categories, with a request-interception safeguard that blocks final side effects and captures screenshots, browser actions, HTTP traffic, session recordings, and agent messages.