ClawBench: Can AI Agents Complete Everyday Online Tasks?
2026-04-09 • Computation and Language
Computation and LanguageArtificial Intelligence
AI summaryⓘ
The authors created ClawBench, a test with 153 everyday online tasks from 15 categories to see how well AI agents can handle real-world activities like shopping, booking appointments, or filling out forms. Unlike other tests, ClawBench uses live websites to keep things realistic and more challenging. They found that even advanced AI models can only complete a small part of these tasks, showing there's still a long way to go before AI can be a reliable all-around helper. Their work helps measure and push forward the development of better AI assistants.
AI agentsautomationbenchmarkweb interactionmulti-step workflowslive websitesform fillingevaluation frameworkgeneral-purpose AIClawBench
Authors
Yuxuan Zhang, Yubo Wang, Yipeng Zhu, Penghui Du, Junwen Miao, Xuan Lu, Wendong Xu, Yunzhuo Hao, Songcheng Cai, Xiaochen Wang, Huaisong Zhang, Xian Wu, Yi Lu, Minyi Lei, Kai Zou, Huifeng Yin, Ping Nie, Liang Chen, Dongfu Jiang, Wenhu Chen, Kelsey R. Allen
Abstract
AI agents may be able to automate your inbox, but can they automate other routine aspects of your life? Everyday online tasks offer a realistic yet unsolved testbed for evaluating the next generation of AI agents. To this end, we introduce ClawBench, an evaluation framework of 153 simple tasks that people need to accomplish regularly in their lives and work, spanning 144 live platforms across 15 categories, from completing purchases and booking appointments to submitting job applications. These tasks require demanding capabilities beyond existing benchmarks, such as obtaining relevant information from user-provided documents, navigating multi-step workflows across diverse platforms, and write-heavy operations like filling in many detailed forms correctly. Unlike existing benchmarks that evaluate agents in offline sandboxes with static pages, ClawBench operates on production websites, preserving the full complexity, dynamic nature, and challenges of real-world web interaction. A lightweight interception layer captures and blocks only the final submission request, ensuring safe evaluation without real-world side effects. Our evaluations of 7 frontier models show that both proprietary and open-source models can complete only a small portion of these tasks. For example, Claude Sonnet 4.6 achieves only 33.3%. Progress on ClawBench brings us closer to AI agents that can function as reliable general-purpose assistants.