Online web agents often augment a base actor with memory, workflow, or skill modules. These modules can improve performance, but they also consume test-time tokens, a cost rarely reported alongside the actor's inference cost. We study online augmentation, where this overhead is paid on every task, and re-evaluate its benefits under a fixed total inference budget. We compare AWM, ASI, and ReasoningBank with a token-matched vanilla baseline that uses the same budget for additional actor steps. Across three WebArena domains and three models, Gemini 3 Flash, GPT-5.4-mini, and Qwen 3.6-27B, the vanilla baseline matches or surpasses all three augmentation methods in aggregate success rate while often using fewer total tokens. We observe a similar trend on WorkArena-L1 with Qwen 3.6-27B, indicating that the effect extends to enterprise knowledge-work tasks. Our results suggest that skills and workflow memory can be useful in specific domains, but their apparent gains often vanish against a budget-matched actor. We further show that run-to-run variance materially affects outcomes and should be reported as a core evaluation criterion for online web agents.

翻译：在线网络智能体通常通过记忆、工作流或技能模块增强基础执行器。这些模块虽能提升性能，但会消耗测试阶段的Token——这一成本在智能体推理开销之外鲜有报告。本研究聚焦于在线增强场景（此类开销在每个任务中均需承担），在固定总推理预算下重新评估其收益。我们将AWM、ASI和ReasoningBank与采用相同预算执行额外智能体步数的Token匹配朴素基线进行对比。在WebArena的三个领域及Gemini 3 Flash、GPT-5.4-mini和Qwen 3.6-27B三种模型上，该朴素基线的总体成功率持平或超越所有三种增强方法，且通常消耗更少的Token总量。在基于Qwen 3.6-27B的WorkArena-L1任务中观察到类似趋势，表明该效应可推广至企业知识工作场景。研究结果表明，技能与工作流记忆在特定领域可能具有实用价值，但其表面增益在预算匹配的智能体面前往往消失。我们进一步证明，运行间方差会实质性影响结果，应作为在线网络智能体的核心评估指标纳入报告。