Claude Managed Agents 新增自托管沙箱与 MCP 隧道
New in Claude Managed Agents: self-hosted sandboxes and MCP tunnelsMay 19, 2026
New in Claude Managed Agents: self-hosted sandboxes and MCP tunnels
Claude 托管代理新功能：自托管沙箱与 MCP 隧道
- Category
- Product announcements
- Product
- Claude Platform
- Date
- May 19, 2026
- Reading time
- 5
- min
- Share
- Copy link
- https://claude.com/blog/claude-managed-agents-updates
- 分类
- 产品公告
- 产品
- Claude 平台
- 日期
- 2026年5月19日
- 阅读时间
- 5
- 分钟
- 分享
- 复制链接
- https://claude.com/blog/claude-managed-agents-updates
Starting today, Claude Managed Agents can operate in a sandbox you control and connect to your private Model Context Protocol (MCP) servers. Both the sandbox where an agent executes tools and the services it reaches run within the established boundaries of your enterprise, under your security and runtime controls.
从今天起，Claude 托管代理可以在您控制的沙箱中运行，并连接到您的私有模型上下文协议（MCP）服务器。代理执行工具所在的沙箱及其访问的服务，都运行在您企业既定的边界内，受您的安全和运行时控制。
The sandbox runs on your own infrastructure, or with managed providers like Cloudflare, Daytona, Modal, or Vercel to handle the compute and isolation for you.
沙箱可以在您自己的基础设施上运行，或使用 Cloudflare、Daytona、Modal 或 Vercel 等托管提供商为您处理计算和隔离。
On the Claude Platform, self-hosted sandboxes is available in public beta and MCP tunnels in research preview (request access).
在 Claude 平台上，自托管沙箱已公开测试版可用，MCP 隧道处于研究预览阶段（需申请访问权限）。
Self-hosted sandboxes: keep agent execution within your perimeter
自托管沙箱：将代理执行保持在您的边界内
A self-hosted sandbox lets a Claude Managed Agent execute tools on infrastructure you control or with a managed sandbox provider. Code execution, sensitive files, packages, services, and data stay within your enterprise perimeter, under your security and runtime controls.
自托管沙箱允许 Claude 托管代理在您控制的基础设施或托管沙箱提供商上执行工具。代码执行、敏感文件、包、服务和数据都保留在您的企业边界内，受您的安全和运行时控制。
With self-hosted sandboxes, you keep sensitive files, packages, and services in your own infrastructure or with a managed sandbox provider. The agent loop that handles orchestration, context management, and error recovery stays on Anthropic’s infrastructure, while tool execution moves to your own configured environment.
使用自托管沙箱，您可以将敏感文件、包和服务保留在您自己的基础设施或托管沙箱提供商中。负责编排、上下文管理和错误恢复的代理循环保留在 Anthropic 的基础设施上，而工具执行则移至您自己配置的环境中。
Inside your perimeter, network policies, audit logging, and security tooling are already in place, and files and repositories don't leave. You also control the compute: resource sizing and the runtime image are set on your side, so agents running compute-heavy work such as long builds or image generation get the CPU, memory, and capacity the task needs.
在您的边界内，网络策略、审计日志和安全工具已经就位，文件和仓库不会外泄。您还可以控制计算资源：资源大小和运行时镜像在您这边设置，因此运行计算密集型工作（如长时间构建或图像生成）的代理可以获得任务所需的 CPU、内存和容量。
Choose your sandbox client
选择您的沙箱客户端
Bring any sandbox client you want, or start with one of our supported providers:
您可以自带任何沙箱客户端，或从我们支持的提供商之一开始：
- Cloudflare runs sandboxes at scale using microVMs and lighter weight isolates. Outbound network requests are in your control with zero-trust secrets injection, customizable proxies to audit, reroute, or modify egress, and the ability to connect to internal services over Cloudflare's network. Amplitude is building Design Agent, an internal tool for on-brand production UI and marketing design, on Managed Agents and Cloudflare for tighter observability and control.
- Daytona sandboxes are full composable computers, long-running and stateful. The same primitive runs a quick burst or an agent that works for hours. The sandbox stays accessible while a session runs over SSH or an authenticated preview URL, or can be paused and restored with full state preserved. Clay’s GTM engineering agent, Sculptor, builds, tests, and monitors workflows autonomously on Managed Agents and Daytona.
- Modal is a cloud platform built for AI workloads, where sandboxes share the same foundation as Modal's functions, storage, and networking primitives, giving you everything you need to build production AI systems. Modal's custom container runtime delivers sub-second startup on any image, scales to hundreds of thousands of concurrent sandboxes, and gives you CPU and GPU resources on demand.
- Vercel sandboxes combine VM security, VPC peering, and bring your own cloud with millisecond startup time. Managed Agents handles the model, tools, and session state, while the Vercel Sandbox firewall injects credentials at the network boundary so they never enter the sandbox. Rogo, an AI platform for institutional finance, is building an analyst agent on Managed Agents and Vercel Sandbox to handle their proprietary data securely.
- Cloudflare 使用微虚拟机（microVMs）和更轻量的隔离环境大规模运行沙箱。出站网络请求由您控制，支持零信任密钥注入、可自定义代理以审计、重路由或修改出口流量，并可通过 Cloudflare 网络连接到内部服务。Amplitude 正在基于 Managed Agents 和 Cloudflare 构建 Design Agent，这是一个用于品牌化生产 UI 和营销设计的内部工具，以实现更紧密的可观测性和控制。
- Daytona沙箱是完全可组合的计算机，长期运行且有状态。同一原语可运行快速突发任务或持续工作数小时的代理。会话通过SSH或经过认证的预览URL运行时，沙箱保持可访问，也可以暂停并在完整保留状态的情况下恢复。Clay的GTM工程代理Sculptor在Managed Agents和Daytona上自主构建、测试和监控工作流。
- Modal是一个为AI工作负载构建的云平台，其沙箱与Modal的函数、存储和网络原语共享同一基础，为您提供构建生产级AI系统所需的一切。Modal的自定义容器运行时可在任何镜像上实现亚秒级启动，可扩展到数十万个并发沙箱，并按需提供CPU和GPU资源。
- Vercel沙箱结合了VM安全性、VPC对等连接和自带云，启动时间仅为毫秒级。Managed Agents处理模型、工具和会话状态，而Vercel沙箱防火墙在网络边界注入凭据，使其永远不会进入沙箱。Rogo是一个面向机构金融的AI平台，正在Managed Agents和Vercel Sandbox上构建分析师代理，以安全处理其专有数据。
MCP tunnels: Connect to services within your private network
MCP隧道：连接到私有网络内的服务
MCP tunnels connect Claude Managed Agents to Model Context Protocol (MCP) servers inside your private network without exposing them to the public internet. Internal databases, private APIs, knowledge bases, and ticketing systems become tools your agents can call. A lightweight gateway you deploy makes a single outbound connection, no inbound firewall rules, no public endpoints, and traffic encrypted end to end.
MCP隧道将Claude Managed Agents连接到私有网络内的Model Context Protocol（MCP）服务器，而无需将其暴露于公共互联网。内部数据库、私有API、知识库和工单系统成为代理可以调用的工具。您部署的轻量级网关仅建立单个出站连接，无需入站防火墙规则，无公共端点，且流量端到端加密。
MCP tunnels is supported in Managed Agents and the Messages API. MCP tunnels is managed from workspace settings within the Claude Console by organization admins.
MCP隧道在Managed Agents和Messages API中受支持。MCP隧道由组织管理员在Claude Console的工作区设置中管理。
“When building Sculptor, Clay's GTM engineering agent for building and testing GTM workflows autonomously, we wanted to give it a more flexible and powerful way to take actions than just tools in a loop. Claude Managed Agents let us replicate the power of a local agent with the reliability, versioning, and background execution of a cloud agent. And running it with our sandboxes, like Daytona, gives us control over the filesystem, so we can mount external file stores and install packages on the fly.”
“在构建Sculptor（Clay的GTM工程代理，用于自主构建和测试GTM工作流）时，我们希望赋予它比循环中的工具更灵活、更强大的行动方式。Claude Managed Agents让我们能够复制本地代理的强大功能，同时具备云代理的可靠性、版本控制和后台执行能力。并且将其与我们的沙箱（如Daytona）一起运行，使我们能够控制文件系统，从而可以挂载外部文件存储并即时安装软件包。”
Ryan Chang, AI Engineering
Ryan Chang，AI工程
“Claude Managed Agents handles the agent loop, Vercel's sandboxes give us an environment we can configure for our workloads. This gives us the option to leverage best-in-class infrastructure while we focus on what compounds for a financial AI platform: depth and breadth of tools and data, and a product surface built for how investors and bankers actually work.”
“Claude托管代理处理代理循环，Vercel的沙箱为我们提供了一个可根据工作负载进行配置的环境。这让我们能够利用一流的底层基础设施，同时专注于金融AI平台的核心价值：工具和数据的深度与广度，以及为投资者和银行家实际工作方式打造的产品界面。”
Strib Walker, Head of Product
Strib Walker，产品负责人
“Our use cases require secure orchestration of internal tools across a complex product surface. Modal's sandbox gives us the security boundary our enterprise customers need, and combining it with Claude Managed Agents gives us a powerful harness without hand-rolling extra complexity. We had a working version up in under a week, raising reliability for our customers.”
“我们的用例需要在复杂的产品界面中安全地编排内部工具。Modal的沙箱提供了企业客户所需的安全边界，与Claude托管代理结合，我们无需手动增加额外复杂性就能获得强大的控制框架。我们在不到一周内就完成了可用版本，提高了客户的可靠性。”
Sai Yandapalli, CTO
Sai Yandapalli，首席技术官
“Claude Managed Agents and Cloudflare let us get the first useful version of our design agent running in two days on infrastructure we already know and trust.”
“Claude托管代理和Cloudflare让我们在两天内就在熟悉且信赖的基础设施上运行了设计代理的首个实用版本。”
Will Newton, Design
Will Newton，设计部门
“As we scale agentic commerce for local businesses, we need a highly efficient path to production with full harness control, scale, and reliability. We're excited to evaluate Claude Managed Agents for this next step, building on our Al infrastructure with Modal!”
“随着我们为本地企业扩展代理商务，我们需要一条高效的生产路径，同时具备完整的控制框架、可扩展性和可靠性。我们期待评估Claude托管代理作为下一步，基于我们与Modal的AI基础设施进行构建！”
Andy Fang, Co-founder
Andy Fang，联合创始人
PrevPrev
上一页
0/5
0/5
NextNext
下一页
eBook
电子书
Getting started
入门指南
Both self-hosted sandboxes and MCP tunnels work within the same core primitives supported by Managed Agents. Self-hosted sandboxes is available in public beta and MCP tunnels in research preview. To get started with MCP tunnels, request access.
自托管沙箱和MCP隧道都基于托管代理支持的核心原语工作。自托管沙箱已公开测试，MCP隧道处于研究预览阶段。要开始使用MCP隧道，请申请访问权限。
Explore our docs to learn more, follow our cookbooks to set up your sandbox provider, or deploy your first agent in the Claude Console.
查阅我们的文档了解更多信息，按照我们的操作指南设置沙箱提供商，或在Claude控制台中部署您的第一个代理。
FAQ
常见问题
No items found.
未找到项目。
Related posts
相关文章
Explore more product news and best practices for teams building with Claude.
探索更多产品新闻和针对使用Claude构建的团队的最佳实践。
Aug 28, 2026
2026年8月28日
Claude for Teachers, now available for U.S. K-12 schools and districts
Claude for Teachers现已面向美国K-12学校和学区开放
Product announcements
产品公告
Claude for Teachers, now available for U.S. K-12 schools and districtsClaude for Teachers, now available for U.S. K-12 schools and districts
Claude for Teachers现已面向美国K-12学校和学区开放Claude for Teachers现已面向美国K-12学校和学区开放
Claude for Teachers, now available for U.S. K-12 schools and districtsClaude for Teachers, now available for U.S. K-12 schools and districts
Claude for Teachers现已面向美国K-12学校和学区开放Claude for Teachers现已面向美国K-12学校和学区开放
Aug 26, 2026
2026年8月26日
Claude gets its own browser in Cowork
Claude 在 Cowork 中获得自己的浏览器
Product announcements
产品公告
Claude gets its own browser in CoworkClaude gets its own browser in Cowork
Claude 在 Cowork 中获得自己的浏览器Claude 在 Cowork 中获得自己的浏览器
Claude gets its own browser in CoworkClaude gets its own browser in Cowork
Claude 在 Cowork 中获得自己的浏览器Claude 在 Cowork 中获得自己的浏览器
Aug 26, 2026
2026年8月26日
Claude in Chrome is generally available
Claude 在 Chrome 中正式可用
Product announcements
产品公告
Claude in Chrome is generally available Claude in Chrome is generally available
Claude 在 Chrome 中正式可用 Claude 在 Chrome 中正式可用
Claude in Chrome is generally available Claude in Chrome is generally available
Claude 在 Chrome 中正式可用 Claude 在 Chrome 中正式可用
Aug 11, 2026
2026年8月11日
Compliance API coverage extends to Claude Cowork and Claude Code
合规 API 覆盖范围扩展至 Claude Cowork 和 Claude Code
Enterprise AI
企业 AI
Compliance API coverage extends to Claude Cowork and Claude CodeCompliance API coverage extends to Claude Cowork and Claude Code
合规 API 覆盖范围扩展至 Claude Cowork 和 Claude Code合规 API 覆盖范围扩展至 Claude Cowork 和 Claude Code
Compliance API coverage extends to Claude Cowork and Claude CodeCompliance API coverage extends to Claude Cowork and Claude Code
合规 API 覆盖范围扩展至 Claude Cowork 和 Claude Code合规 API 覆盖范围扩展至 Claude Cowork 和 Claude Code
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力