- 类别产品公告
- 产品Claude 平台
- 日期2026 年 6 月 9 日
- 阅读时间5分钟
- https://claude.com/blog/whats-new-in-claude-managed-agents
从今天开始，Claude 托管智能体可以按计划运行，并安全访问 CLI 工具及其他经过身份验证的服务。这两项功能现已在 Claude 平台上以公开测试版形式提供。
计划部署：按计划运行智能体
智能体现在可以按计划运行，自动完成常规工作。计划部署会按照 cron 计划运行一个 Claude 托管智能体。每次计划触发时，智能体都会启动一个新会话并完成其任务，无需您自行构建或托管调度器。
可用于重复性工作，例如夜间数据同步、每周合规扫描或每日摘要。部署上线后，您可以随时暂停、恢复或归档，也可以按需触发额外运行。
团队已经在使用计划部署来自动化重复性工作：
- 乐天使用计划部署来分析电子表格数据，并按周或按月生成报告和演示文稿。团队还监控生产日志和指标，使产品经理无需创建仪表板即可了解应用运行状况。
- Actively AI 使用托管智能体为销售团队提供跨账户的智能体搜索功能。计划部署会定期刷新答案，通过替换团队最初自行构建的调度基础设施来简化其技术栈。
- Ando 使用计划部署来保持招聘和销售团队的运转。智能体自主监控渠道中提出的下一步行动建议，在到期时进行跟进，并发送会议提醒。
保险库：存储环境变量以对 CLI 及其他工具进行身份验证
保险库可安全存储 Claude 托管智能体的环境变量和凭证。在运行时，智能体将 API 密钥等机密信息作为环境变量进行访问，从而无需将凭证硬编码到提示词或代码中，即可对 CLI 工具和其他服务进行身份验证。
智能体通过直接 API 调用、CLI 和 MCP 连接到外部系统。CLI 让智能体能够通过 shell 直接驱动现有的命令行工具，从而提供了一条快速、轻量的集成路径。将 API 密钥与环境变量名称及其可访问的域名进行注册后，安装在智能体沙箱中的 CLI 便可以使用该密钥进行经过身份验证的 API 调用。
智能体永远不会看到你的密钥，因为沙箱中只保存了一个占位符。真正的密钥在网络边界处附加，并且仅在你允许的域名请求上附加，因此它只会流向你批准的地方。要更改密钥，只需在保管库中更新它，正在运行的会话将在下一次调用时获取新值。大多数通过 HTTP 请求发送密钥的 CLI 都采用这种方式工作，包括 Browserbase、KERNEL、Notion、Ramp 和 Sentry 的 CLI。Browserbase 和 KERNEL 首次为托管智能体提供了浏览器能力，使得智能体能够与其他工具一起浏览和交互网页。
团队正在使用保管库中的环境变量，为智能体提供对经过身份验证的工具的安全访问：
- Notion 使用保管库中的环境变量来部署其 CLI 以及 MCP 工具，为其智能体增加了文件上传功能，而无需将 API 令牌交给模型。
- Browserbase 使用 browse CLI 构建了其公开的浏览器技能目录，并通过保管库进行身份验证。定期部署会周期性地验证该目录以保持其准确性。
- KERNEL 使用保管库中的环境变量，将智能体安全地连接到用于跟踪使用情况和客户对话的数据库。智能体会在用量激增发生时立即标记出来，以便团队与客户确认该活动是否属于预期行为。
- Milana 使用保管库中的环境变量，将其 AI 产品工程师安全地连接到客户的代码库。该智能体会自动发现并修复错误，大规模数据分析的运行速度也比以前更快。
“保管库中的环境变量让我们能够安全地部署 Notion CLI，通过确保敏感的 API 令牌永远不会交给智能体，来满足我们安全团队的严格指导方针。该 CLI 是对 MCP 工具的补充，能够在 Claude 托管智能体中实现文件上传功能。”
Quan Nguyen，公共 API 负责人
“乐天多个团队使用定时部署来分析电子表格中的数据，并按周或按月生成报告或演示文稿。我们的高级用户将其应用于生产日志和指标，这样产品经理无需创建分析仪表板，就能了解其应用的运行状况。”
Yusuke Kaji，AI 业务总经理
“我们的大多数用户更倾向于使用较少的智能体，而非大量智能体。通过定时部署，他们可以将更多能力整合到一个自主智能体中。例如，一个智能体可以同时监控多个销售和招聘流程，与相关人员核实最新进展，并推动后续步骤。”
Sara Du，创始人
“我们在 Claude 托管智能体上构建了一个跨账户的智能体搜索系统，让销售团队可以询问诸如‘今天应该联系哪些账户’之类的问题。由于客户希望这些答案能定期更新，我们用定时部署取代了自己构建的调度基础设施，这极大地简化了我们的技术栈，并改善了我们的产品周期。”
Mihir Garimella，联合创始人
“通过 Claude 托管智能体，我们将智能体锚定在客户的实际代码库上。我们发现它易于使用，并且几乎立即获得了高质量的结果。关键突破在于保险库中的环境变量，它让我们的智能体能够通过 CLI 调用私有 API，而无需暴露凭证。大规模数据分析现在变得极其迅速，并且借助输出结果，我们可以确保每个输出的质量。”
Raghav Sethi，联合创始人兼首席技术官
“保险库中的环境变量使我们的工程团队能够结合两种主要的计算原语：智能体和浏览器。在 Browserbase，我们使用带有浏览 CLI 的 Claude 托管智能体来生成我们的公开浏览器技能目录，这些技能帮助智能体浏览网页，而定时部署则对我们的公开目录进行定期验证。”
Ziray Hao，产品负责人
在 Kernel 的浏览器基础设施上，使用量可能会迅速激增，通常就在客户部署之后。借助保管库中的环境变量，我们的智能体现在可以直接连接到我们追踪使用量和客户对话的数据库。它能在几秒内拉取过去 30 天的每日使用数据，在激增发生时立即标记出来，并帮助我们的团队向客户确认该活动是预期行为。
Catherine Jue，联合创始人兼首席执行官
开始使用
查阅我们的文档以了解更多信息，或访问 Claude 控制台来部署你的第一个智能体。
借助 Claude 改变你组织的运作方式
- Category
- ProductClaude Platform
- DateJune 9, 2026
- Reading time5min
- https://claude.com/blog/whats-new-in-claude-managed-agents
Starting today, Claude Managed Agents can run on a schedule and securely access CLI tools and other authenticated services. Both features are now available in public beta on the Claude Platform.
Scheduled deployments: Run agents on a schedule
Agents can now run on a schedule, completing routine work automatically. A scheduled deployment runs a Claude Managed Agent on a cron schedule. Each time the schedule fires, the agent starts a new session and completes its task, with no scheduler for you to build or host.
Use it for recurring work like a nightly data sync, a weekly compliance scan, or a daily digest. Once a deployment is live, you can pause, resume, or archive it at any time, or trigger additional runs on demand.
Teams are already using scheduled deployments to automate recurring work:
- Rakuten uses scheduled deployments to analyze spreadsheet data and produce reports and decks on a weekly or monthly schedule. Teams also monitor production logs and metrics, allowing product managers to see application health without creating a dashboard.
- Actively AI uses Managed Agents to power cross-account agentic search for sales teams. Scheduled deployments refresh answers regularly, simplifying their stack by replacing scheduling infrastructure the team initially built themselves.
- Ando uses scheduled deployments to keep hiring and sales teams moving. Agents autonomously watch channels for proposed next steps, follow up when they're due, and send meeting reminders.
Vaults: store environment variables to authenticate CLIs and other tools
Vaults securely store environment variables and credentials for Claude Managed Agents. At runtime, agents access secrets like API keys as environment variables, so they can authenticate CLI tools and other services without hardcoding credentials into prompts or code.
Agents connect to external systems through direct API calls, CLIs, and MCP. CLIs let agents drive existing command-line tools directly through a shell, making them a fast, lightweight integration path. Register an API key with an environment variable name and the domains it can reach, and the CLIs installed in an agent's sandbox can use it to make authenticated API calls.
The agent never sees your key because the sandbox only holds a placeholder. The real key is attached at the network boundary, and only on requests to domains you allow, so it only goes where you’ve approved. To change a key, update it in the vault, and running sessions will pick up the new value on their next call. Most CLIs that send their key in an HTTP request work this way, including the Browserbase, KERNEL, Notion, Ramp, and Sentry CLIs. Browserbase and KERNEL give Managed Agents browser capabilities for the first time, so agents can navigate and interact with the web alongside their other tools.
Teams are using environment variables in vaults to give agents secure access to authenticated tools:
- Notion uses environment variables in vaults to roll out its CLI alongside MCP tools, adding file-upload capabilities to its agents without API tokens ever being handed to the model.
- Browserbase built its public catalog of browser skills using the browse CLI, authenticated through vaults. A scheduled deployment periodically validates the catalog to keep it accurate.
- KERNEL uses environment variables in vaults to securely connect agents to the databases where it tracks usage and customer conversations. The agent flags usage surges as they happen, so the team can confirm with customers if the activity is intended.
- Milana uses environment variables in vaults to securely connect its AI product engineer to a customer's codebase. The agent finds and fixes bugs automatically, with large-scale data analysis running faster than before.
“Environment variables in vaults let us securely roll out the Notion CLI, meeting our security team’s strict guidelines by ensuring sensitive API tokens are never handed to agents. The CLI is complementary to MCP tools, enabling file-upload capabilities in Claude Managed Agents.”
Quan Nguyen, Public API Lead
“Teams across Rakuten use scheduled deployments to analyze data in a spreadsheet and produce a report or deck on a weekly or monthly schedule. Our power users put it on production logs and metrics, so a product manager can see the health of their application without creating an analytics dashboard.”
Yusuke Kaji, General Manager of AI for Business
“Most of our users prefer to work with fewer agents rather than many. With scheduled deployments, they can bundle more capabilities into one autonomous agent. For example, an agent can watch multiple sales and hiring processes, check in with the right people for updates, and push next steps along.”
Sara Du, Founder
“We built a cross-account agentic search system on Claude Managed Agents that lets sales teams ask things like which accounts to reach out to today. Since customers want those answers refreshed regularly, we replaced the scheduling infrastructure we'd built ourselves with scheduled deployments, which greatly simplified our stack and improved our product cycles.”
Mihir Garimella, Co-founder
“With Claude Managed Agents, we grounded our agent on a customer's actual codebase. We found it easy to work with and got high-quality results almost instantly. The key unlock was environment variables in vaults, which let our agent invoke private APIs through a CLI without exposing credentials. Large-scale data analysis is now dramatically faster, and with outcomes we can ensure the quality of every output.”
Raghav Sethi, Co-founder & CTO
“Environment variables in vaults enabled our engineering team to combine two major compute primitives: the agent and the browser. At Browserbase, we used Claude Managed Agents with the browse CLI to generate our public catalog of browser skills that help agents navigate the web, and scheduled deployments run periodic validation on our public catalog.”
Ziray Hao, Product Lead
Usage on Kernel's browser infrastructure can surge quickly, often right after a customer deploys. With environment variables in vaults, our agent now connects directly to the databases where we track usage and customer conversations. It pulls 30 days of daily usage in seconds, flags surges as they happen, and helps our team confirm with customers that the activity is intended.
Catherine Jue, Co-founder & CEO
Getting started
Explore our documentation to learn more or visit the Claude Console to deploy your first agent.