Anthropic 推出自己的个人Agent： Claude Duet 能主动跟进工作，电脑合上也能继续干活
它会查看已连接的应用、准备简报和工作结果，再把需要你决定的事送到面前；对外发送、付款、签名和删除仍要先获批准。
Anthropic 推出了 Claude Duet，这是一款会主动跟进工作、替你执行任务的独立 Agent 应用。
主要功能特点包括：
- 主动跟进与清晨简报：每天清晨自动梳理各关联应用的关键动态并生成摘要；定期检查已连接应用，有需要时才发信通知。
- 离开设备后持续运转：任务运行在 Anthropic 托管的独立云端虚拟机与浏览器中，手机关机或电脑合盖都不会打断正在推进的工作。
- 高风险动作人工确认：自主完成跨软件的信息查阅、分析和草稿拟定，但在对外发邮件、结算商户款项、签署文件或删除数据前，必须停下等待你批准。
- 承接既有 Claude 记忆与技能：直接继承原有 Claude 账号中的记忆、技能与已连接应用，在同一条持续对话中（支持打字与语音）统一管理即时吩咐与长周期任务。
约 1 分 39 秒，保留原片与原声；前半段展示产品流程，后半段是 Anthropic 员工的使用反馈。画面中的业务、金额与结果属于演示，不是实测成绩。
它会先发现需要处理的事
使用普通 AI 聊天时，通常由人先发现问题并发起对话。Claude Duet 则会主动带着需要处理的事来找你，并在你离开时继续把事情做完。
Duet 以 Claude Opus 5.5 为核心，判断哪些事项可以等待、哪些可以先准备、哪些需要请你决定。
根据 Anthropic 公布的早期测试者反馈，这种主动性已经在一些业务场景中有所体现：
- 日程疏漏的事先提醒：房产经纪与机构负责人 Jess 每天早晨会收到 Duet 整理的当日简报。某天 Duet 在日常检查日程表时，主动捕捉到一场房产带看被排在错误的日期，并及时向她预警。Jess 表示这正是她希望 AI 主动发现、而不需要自己专门发问的事务。
- 多业务系统的信息整合：南佛罗里达拥有约 110 名员工的餐饮店管理者 Amin，将 Duet 用于会计、客服与营销工作，并逐店核对发票。他表示 Duet 把这些事务整合在一起，“我不用再打开 100 个标签页了”。
- 行程冲突的预先提议：品牌策略师 Mya 的日程表上排定了一场与潜在客户的电话会议。Duet 为她整理好了沟通背景简报，同时注意到这场通话紧挨着她的航班，便主动提议调整会议时间。
Mya 收到的是改期建议；是否调整安排，仍由她决定。
为什么断网关机后工作仍在推进
关键在于它的计算环境并非运行在你的本地手机或笔记本电脑上。当你在 Duet 中交代任务后，工作由云端虚拟机及其自带的独立浏览器接管推进：
- Opus 5.5 负责判断步骤：理解业务目标，规划执行步骤，决定何时调用工具、何时需要停下请示。
- 云端工具负责具体执行：在云端虚拟机和独立浏览器中调取应用数据、处理具体操作，准备文稿、图表、仪表盘、财务预测或网页应用。
手机与电脑只是下达任务与接收关键通知的窗口。即便手机关机、电脑合盖或暂时离线，云端虚拟机仍会在后台继续推进任务；等到任务完成或需要你的授权时，设备恢复联网就能收到通知。
主动检查不等于无节制的实时盯守
Duet 的主动性并不是在后台无休止地实时盯守。
官方说明中，Duet 每隔几小时检查一次已连接的应用。之所以采用这种间隔，主要是因为不停检查会更快耗尽订阅计划的用量限额。至于何时向你发消息，则由模型的判断能力来决定什么需要你关注、什么可以等待，Anthropic 也在根据测试反馈持续调整这套判断。
这也意味着，如果突发变动或紧急消息刚好落在两次检查之间，Duet 可能会遗漏。你可以直接告诉 Duet 自己最关心的内容以及期望接收通知的频次。
授权分流：哪些事自主做，哪些事绝不碰
将 AI 连入企业核心工具后，最大的担忧往往是它会不会擅自操作或泄露数据。Claude Duet 对不同动作与数据设定了清晰的权限边界。
以“跟进客户欠款”为例：整理账单和拟稿可以先做，真正发送提醒要等你批准。
读取已连接应用、查未付发票、分析数据、起草提醒。
发送消息、付款、签署文件、删除内容，都要先展示并等待确认。
账户之间转移资金、证券交易，仍属于禁止的操作。
信用卡号、政府身份信息：新分类器与安全提示词旨在拒绝外发请求。
健康、个人财务信息：只有你明确要求发送，才允许放行。
按官方规则整理的示意，不代表产品实际界面或内部执行顺序。
1. 行动的权限分级
- 自主执行：自主读取信息、交叉分析并拟定草稿或准备方案。
- 需要人工批准：在向外发送消息、支付款项、签署文件或删除任何内容前，Duet 会展示即将执行的具体操作并等待你批准。例如：
- 发送邮件或 Slack 消息；
- 签署文件或合同；
- 删除内容；
- 采购付款：使用 Stripe 的 Link 钱包时，Duet 会先与你确认商品，再由 Link 弹出付款审批，并通过虚拟卡完成支付，避免暴露真实卡号；此外也支持通过 Shopify 的 Shop Pay 在线结账。
- 坚决禁止的操作：有些行为即使获得使用者授权，Duet 也绝不执行，包括账户间资金划转和证券交易。商户结账属于采购付款，与账户间直接转账是完全不同的两类操作。
2. 敏感信息的防护机制
Agent 在替你阅读资料与拟定外发沟通时，需要不断决定信息流向。Duet 引入了专属的新分类器和安全系统提示词来检查每条消息，防范敏感信息外泄：
- 最高敏感级别信息：旨在直接拒绝发送信用卡号和政府颁发的身份证明等信息。
- 其他敏感信息：医疗健康数据或个人财务状况等，只有在使用者明确要求发送时才会被允许放行。
3. 凭据隔离与安全防护
在连接企业应用后的凭据保管上，Duet 沿用了与 Claude Managed Agents 相同的设计思路，模型本身不会直接接触你的登录密码或登录令牌。Managed Agents 的做法是：密码和登录凭证由执行环境之外的设施保管，Agent 调用工具时，由中间服务完成身份验证，模型无需拿到密码。
此外，Duet 还设置了针对性的防御机制：
- 防范提示注入：吸纳了浏览器使用（browser-use）等相关研究的防御成果，防范外部网页或邮件中夹带隐藏指令，诱使 Agent 偏离目标或越权行事；
- 动作分类器（Auto mode classifier）：防止 Duet 擅自执行使用者未吩咐的操作步骤。
Anthropic 开展了红队对抗测试、聘请外部安全公司进行了渗透测试，并将 Duet 纳入公开漏洞赏金计划。这些措施用于降低越权和泄露风险，实际使用仍需检查重要结果。
应用连接器的真实边界
Claude Duet 可以调用 Claude Marketplace 中提供的超过 6,000 个连接器与插件。
涵盖 Gmail、Google Calendar、Slack、Shopify、QuickBooks、Gusto、HubSpot 等主流办公与商业系统；图中包含 1Password 图标，但官方确认其保存登录凭据的支持仍在开发中，尚未正式上线。
连接器的作用是让 Agent 在获得授权后使用对应的外部应用。“6,000 多个”统计的是市场中的连接器与插件，不能据此判断每款软件的全部操作是否都能顺利完成。
在具体业务场景中，Duet 可以支持多条典型工作流：
- 未结发票与催款：检索 Intuit QuickBooks 或 Stripe 中的未付发票并拟定催款提醒，经你审核批准后再对外发出；
- 工资与人事核算：在 Gusto 中填报工时、奖金和报销款项，准备好工资单供人工审核发放；
- 客户跟进与签约：利用 HubSpot 中的客户与交易历史拟定跟进邮件；在 LinkedIn 起草招聘文案并在批准后发布，或通过 DocuSign 准备合同以供签署；
- 营销设计与电商店铺：使用 Canva 将新品发布转化为社交推广素材；在 Shopify 中搭建与管理店铺。官方说明更丰富的 Shopify 可视化编辑器即将推出，支持在正式上线前预览页面。
此外，官方确认在获得使用者许可的前提下，通过 1Password 中保存的登录信息自动登录网站的支持也即将推出。使用者也可以随时断开对任何单个应用的授权。
开通门槛、可用平台与额度消耗规则
对于打算体验的个人专业人士与小企业主，目前需要了解以下使用规则：
1. 开通资格与平台支持
- 当前阶段：处于研究预览（Research Preview）阶段，必须通过官方候补名单排队申请开通。
- 支持计划：首批仅面向美国地区的 Claude Pro 和 Claude Max 个人订阅用户开放。Team 和 Enterprise 方案目前暂未包含在此次预览范围内。
- 使用载体：支持 iOS 与 Android 的独立原生应用程序；在电脑端，则通过 claude.ai 网页版访问。
2. 额度消耗机制
在研究预览阶段，Claude Duet 包含在 Pro 与 Max 订阅计划中，并与 Claude 的其他功能共享用量限额（Usage limits）。它的额度消耗方式与普通对话不同：
- 后台主动检查计入用量：Duet 在后台定期检查已连接应用本身就会计入用量。
- 复杂大任务涉及多轮工作：当它承接构建财务预测模型或制作演示文稿等较大型任务时，需要在后台触发多轮模型交互与工具操作，因而会占用更多额度。
通知偏好控制你希望听到什么、多久收到消息；后台检查本身仍计入用量。
先从结果容易核对的任务开始
作为早期的研究预览版本，现阶段的 Claude Duet 仍存在一些现实局限：
- 可能误读与误判优先级：它可能会误读长对话上下文，或对事务的紧急程度校准出现偏差，也可能漏掉两次主动检查之间发生的新变动。
- 核心功能围绕工作场景构建：主要针对工作事务设计，处理订票等个人跑腿杂务的能力相对较弱。
- 暂无法查看其自带浏览器：目前 Duet 还不能向你直接展示它自带的浏览器画面，不过使用者依然可以查看它执行的任务步骤与生成的具体结果。
在实际使用中，Duet 是否真正省事，取决于漏项与误判的发生概率，以及人工审核确认各项操作所需的时间与精力成本。在早期体验时，更适合先从范围清晰、容易核验结果的工作任务切入。
Anthropic 推出自己的个人Agent： Claude Duet 能主动跟进工作，电脑合上也能继续干活
它会先发现待办、准备结果，再把需要你决定的事交回来。
独立个人 Agent · 云端执行 · 重要动作先批准
Anthropic 推出了 Claude Duet，一款会主动跟进工作、替你执行任务的个人 Agent。它继承 Claude 的记忆、技能和已连接应用，用同一条文字或语音对话接住日常吩咐与长期任务。
你离开后，它把工作推进到哪一步？
以“跟进未付发票”为例，查账与准备可以先做，真正发送仍要你点头。
从 QuickBooks 或 Stripe 查账单，找出需要跟进的事。
结合查到的信息准备材料，留下可审阅的结果。
展示即将执行的操作，等待你批准发送。
为什么合盖也能继续？Opus 5.5 判断步骤，独立云端虚拟机和浏览器负责执行。手机或电脑离线，云端工作仍可继续；设备重新联网后接收通知。
准备、执行、禁止，分清三类事
查阅资料、分析数据、拟定草稿。
发送消息、付款、签署文件、删除内容。
账户间转账、证券交易。
商户付款与账户间转账规则不同：Link 在商品确认和消费批准后使用虚拟卡付款；Shop Pay 另支持商户结账。
主动跟进，也有检查间隙与额度成本
它并非实时盯守，间隙可能漏看变化；不停检查会更快耗尽额度。后台检查和复杂任务，都计入与其他 Claude 功能共享的 Pro/Max 用量限额。
价值在于先替人发现疏漏
Jess 的带看日程
Duet 主动发现带看排错了日期，她不必先意识到问题再询问。
Mya 的客户通话
它准备客户简报，发现通话紧挨航班并建议改期；是否调整仍由人决定。
现在谁能用，先交给它什么？
美国 Pro/Max 个人用户可加入研究预览候补名单，Team/Enterprise 尚未包含。它可能误读上下文或误判紧急程度，暂不能展示自己的浏览器。
Introducing Claude Duet
Claude Duet is a new app where Claude works as a proactive agent. The research preview opens by waitlist today for Pro and Max subscribers in the US.
Today, we’re opening a research preview of Claude Duet, a new app where Claude works as a proactive agent. It comes to you with what needs your attention, completes tasks you ask of it, keeps working even when you step away, and checks with you before it acts on your behalf. We built Claude Duet for professionals and small business owners with more on their plates than they can track.
It’s available as a standalone app on iOS and Android, and on claude.ai. For this research preview, Claude Duet is included in Pro and Max plans and uses the same usage limits as the rest of Claude. To get access, join the waitlist (opens in new tab).
Put your attention where it matters most
We designed Claude Duet to be a partner you can trust. Professionals of all kinds have to juggle many decisions a day, and we think they’ll find Claude Duet most useful. It helps you prioritize where to spend your time and energy, and to leave room for the ambitious projects that felt impossible or daunting before.
What makes Claude Duet so helpful is its judgment. We designed it around Claude Opus 5.5 (opens in new tab) and built it to be resourceful and to decide what needs you, what can wait, and when to check with you first.
Each morning, it sends you a short brief on the most important items across your connected apps. Throughout the day, it keeps tabs on those apps and messages you only when something needs your attention. You can also hand it tasks of all shapes—from simple administrative tasks of scheduling meetings to more complex ones, such as building dashboards with dozens of tools, creating financial or sales forecasts, prototyping, and building web apps.
Safety and discretion are built in
Claude Duet draws on years of security, safety, and privacy work on agents, both our own and the infrastructure businesses use to run theirs:
- Claude Duet draws on our framework for trustworthy agents (opens in new tab), which also guides agents like Claude Code and Claude in Chrome.
- Claude Duet’s handling of credentials follows the design we use for Claude Managed Agents (opens in new tab), so it never directly sees your passwords or sign-in tokens.
- Actions are approved by our auto mode classifier (opens in new tab), the same safeguard that protects Claude and Claude Code from taking steps you didn’t ask for.
- Claude Duet’s defenses against hidden instructions and prompt injection come from our browser-use research (opens in new tab), and our models and safeguards also cover this more broadly.
- We run adversarial tests for Claude Duet’s security infrastructure. Red teams and an external firm penetration tested Claude Duet, and it’s covered by our public bug bounty program (opens in new tab).
How Duet handles sensitive information. We’re also releasing Claude Duet with a new dedicated classifier and safety system prompt that checks each message and guards against leaks of sensitive information. An agent that reads email and sends messages for you is constantly deciding where information should go. This new classifier and prompt are designed to reject requests to share the strictest tier of information, including credit card numbers and government identification. Other sensitive information, such as health or personal finances, are let through only when you’ve explicitly instructed Claude Duet to send it.
We’re continuing to develop and strengthen these protections, and will share more details about how Claude Duet is designed for safety in the coming days.
How testers are using Claude Duet
People at Anthropic have been using Claude Duet to accomplish more, but also to leave more room for thinking and being present. Boris Cherny, who leads our Claude Code team, talks through his mental to-do list with Claude Duet on walks, instead of saving the work for his desk. Chelsea Larsson, who leads our model and product behavior design team, hands Claude Duet her to-do list before leaving to be with her kids and it saves the results for an evening report. And the team behind Claude Duet uses it to build the product. Karan Ahuja, who leads the team, works with it to schedule user interviews, partner with him on prototypes, and even ship code.
We gave small business owners and entrepreneurs early access to Claude Duet, and some have been using it to help run their businesses. Here’s what some of them said:
How Claude Duet works
Your context from Claude is the foundation. Claude Duet carries over your memory, skills, and connected apps from your Claude account, so setting it up requires minimal work. What changes is how you work together: Claude Duet comes to you first, and you can message it through typing or voice in one ongoing thread, so a quick reply and a weeklong project both start in the same place.
Work continues when you step away. Claude Duet has a virtual machine with its own browser, so it doesn’t require your laptop or phone to be on. Describe a task in a short message, and it works out which apps to use and what steps to take in the background. You’ll receive a notification when the task is complete or needs your review.
Receive a message when something needs you. Claude Duet checks your connected accounts every few hours, because checking nonstop would use up your plan’s limits much faster. We’re continuing to test its judgment about when to message you, and we’ll keep tuning it based on what testers tell us. You can also tell it what you want to hear about and how often.
Important decisions stay with you. Claude Duet reads, drafts, and prepares work on its own. But before it sends a message, makes a payment, signs a document, or deletes anything, it shows you exactly what it’s about to do and waits for your approval. A few things are off-limits even with your approval, like trading securities or moving money between accounts. You’re also able to disconnect any app Claude Duet has access to.
Connect with the tools your business runs on. Claude Duet can use the over 6,000 connectors and plugins in the Claude Marketplace (opens in new tab) to:
- Find the Gmail and Slack messages you’ve missed and help you reply.
- Check Intuit QuickBooks or Stripe for unpaid invoices and draft reminders.
- Prepare and run payroll in Gusto, with hours, bonuses, and reimbursements filled in.
- Draft a job posting and, once you edit and approve, post it to LinkedIn.
- Help prepare agreements and send them for review and signature through Docusign.
- Draft follow-up emails using the contact and deal history in HubSpot.
- Turn your product launch into a social campaign in Canva.
- Set up, design, and manage your Shopify store. Coming soon, a richer visual editor will let you design your storefront and preview it before anything goes live.
You can also use Claude Duet to purchase things for your business using Link, Stripe’s digital wallet. Claude Duet confirms the item with you, and Link asks you to approve the spend. It then pays with a virtual card, so your real card number isn’t shared. It can also use Shop Pay from Shopify to check out at online stores. Coming soon, with your permission, it will be able to sign in to sites with logins you’ve saved in 1Password.
Join the research preview
This is an early version of Claude Duet, and it still might misread a thread or miscalibrate what’s urgent. We built its core functionality around work, so personal errands, such as booking tickets, aren’t quite as strong, and it can’t yet show you its browser. For now, Claude Duet lives in its own app, separate from the main Claude app. What we learn from early users will shape how the two work together.
During the research preview, Claude Duet shares your plan’s usage limits with the rest of Claude, and it uses them differently than a typical conversation does. Its background checks count toward those limits, and larger tasks, like building a financial model or a deck, require more model turns and background work.
We’re starting today with Pro and Max users in the US. Join the waitlist (opens in new tab) and we’ll email as soon as Claude Duet’s ready for you.