Capsulate

io.github.capsulatedev由 GitHub 账号 capsulatedev 发布v0.7.0更新于 Oct 24, 2026

Real Linux and Windows desktop VMs for AI agents: exec, files, computer-use and snapshots via MCP.

已验证STDIO仅桌面云端与基础设施浏览器自动化开发者工具

概览

AI 生成的概览

Capsulate 让 AI 助手启动并操作隔离的云端 Linux 和 Windows 桌面：执行命令、读写文件、使用图形界面、浏览网页并创建快照。

可以做什么
启动、列出、查看、快照和销毁云端胶囊，并通过目录工具发现可启动的模板与套件。执行 shell 命令、python/node/bash 片段，读写文件并设置环境变量。控制桌面：截图、鼠标与键盘输入、无障碍树、元素查找与点击、窗口、剪贴板、OCR 以及批量动作序列。通过页面结构驱动胶囊自带的 Chrome，进行导航、标签页、点击、填写、求值与截图。通过签名控制台链接共享实时屏幕，配合旁白、控制权交接与交回。学习、保存、审批、演练并运行团队技能，并查看待处理运行的通知收件箱。
适用场景
适合需要一台真实、可丢弃的机器而非本地工具的场合：在干净环境中运行测试套件或构建、自动化桌面应用，或进行可能需要人工观看并接管的浏览器操作。
运行要求
通过 stdio 在本地运行，通常使用 npx 和 Node.js。需要 Capsulate 账户和 cap_live_ API 密钥，通过 CAPSULATE_API_KEY 提供（也可使用密钥文件）。CAPSULATE_BASE_URL 和 CAPSULATE_TEAM_ID 为可选项。需要访问 Capsulate API 的网络连接。
安装前请注意
API 密钥是可计费的凭据，尽量不放入共享配置文件。胶囊是运行期间产生费用的云端机器，可通过 CAPSULATE_TEAM_ID 指定计费团队。工具会执行命令、写入文件、控制桌面和浏览器并销毁胶囊，操作会改变远端状态。控制台链接虽经签名，但在过期前无需登录即可打开实时屏幕。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Capsulate，将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Desktop only，通过 STDIO。 STDIO 服务会启动本地进程，因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

@capsulate/mcp

The Capsulate Model Context Protocol server. Give Claude Code, Claude Desktop, Cursor, Windsurf, Antigravity, any MCP client, the ability to launch and drive isolated cloud computers ("capsules"), including real Linux and Windows desktops that a person can watch live and take over.

Setup

  1. Create an API key in the Capsulate dashboard: Settings → API & Keys.
  2. Add the server to your MCP client config.

Claude Code

bash
claude mcp add capsulate -e CAPSULATE_API_KEY=cap_live_... -- npx -y @capsulate/mcp

Claude Desktop / Cursor

json
{  "mcpServers": {    "capsulate": {      "command": "npx",      "args": ["-y", "@capsulate/mcp"],      "env": {        "CAPSULATE_API_KEY": "cap_live_…"      }    }  }}

To keep the key out of the config file, put it in a file and point at it with CAPSULATE_API_KEY_FILE instead. The file holds either the bare key or a CAPSULATE_API_KEY=cap_live_... line.

For a self-hosted or non-default deployment, also set "CAPSULATE_BASE_URL": "https://api.your-domain".

Tools

ToolWhat it does
capsulate_launchLaunch a capsule and wait until ready
capsulate_list / capsulate_getInspect capsules
capsulate_destroyTear a capsule down
capsulate_execRun a command (argv array)
capsulate_run_codeRun a python/node/bash snippet
capsulate_write_file / capsulate_read_fileFile I/O
capsulate_set_envSet environment variables
capsulate_snapshotSnapshot a capsule
capsulate_catalogDiscover launchable templates & bundles

Desktop computer use

ToolWhat it does
capsulate_screenshotSee the screen (returns an image; format: "jpeg", a scale or a region make it smaller)
capsulate_screen_infoThe screen's size, the pointer and the active window
capsulate_click / capsulate_move / capsulate_drag / capsulate_scrollPointer
capsulate_type / capsulate_keyKeyboard
capsulate_ui_treeRead the accessibility tree (AT-SPI on Linux, UI Automation on Windows) with clickable centers
capsulate_find_element / capsulate_wait_elementFind or wait for an element by role, name or text
capsulate_click_elementFind an element and click it: the reliable way to press a named button
capsulate_clipboardRead or set the clipboard (paste long text instead of typing it)
capsulate_elementAct on an element without coordinates: set its value, clear, toggle, expand or collapse, invoke, select an option by name, focus, scroll it into view
capsulate_windowsList windows or bring one to the front
capsulate_windowMinimize, maximize, restore, close, move or resize a window
capsulate_openOpen a URL, a file or an app by name, and wait for its window
capsulate_wait_forWait for an element to appear, go away, be enabled or show a text, for a window, or for the screen to settle
capsulate_ocrRead the text on screen when the UI tree has no labels
capsulate_run_actionsSeveral steps in one call, back to back, stopping at the first that fails, with a look at the end

The input verbs accept screenshot: true (and screenshotDelayMs) to return the frame after the action. Every tool works the same on Linux and Windows desktops; a multi-line editor is a document on Windows and a text on Linux.

Browser fast path

The capsule's own Chrome, driven through the page's structure at page speed, on the same screen the person watches: capsulate_browser_open, _navigate, _tabs, _snapshot (an outline with refs), _click, _fill (the value is never echoed), _select, _press, _scroll, _wait, _text, _eval and _screenshot.

Live: a person watches, takes over, hands back

A desktop capsule has one shared screen. The AI drives it through the tools above while a person watches the same screen in a browser, sees the AI's pointer, clicks and keys drawn over it, reads its narration, and can take the screen at any time.

ToolWhat it does
capsulate_console_linkA signed, expiring URL that opens the live screen in any browser with no login (control or view)
capsulate_sayShow one line of narration on the live console
capsulate_controlWho is driving: free, agent (the AI) or human; take or release the screen; set the pace (natural or presentation) for a person watching
capsulate_handoffHand the screen to the person with a note ("Please sign in"), wait until they hand it back, get their note
capsulate_wait_for_handbackKeep waiting for the person to hand the screen back

While a person holds the screen, the AI's input verbs answer human_in_control and nothing reaches the capsule; the tool result says to wait with capsulate_wait_for_handback. Reading the screen (screenshot, UI tree, OCR) keeps working. Each AI input holds the screen for 45 seconds; the console is view-only for everyone else during that time and turns interactive again when the AI goes quiet or releases.

Skills: learn once, replay in minutes

ToolWhat it does
capsulate_lessonLearn a job on a capsule: a person teaches on the shared screen (clicks, keys and typing with the elements they hit, passwords masked, optionally video and voice) or the AI teaches itself toward a goal; get reads the digest
capsulate_skillsThe team's skills: save one from a lesson's digest, versions, approve, save an environment, rehearse, weekly rehearsals
capsulate_skill_runRun a skill's procedure in the background; it asks you where the skill leaves a decision open
capsulate_inboxRuns that wait, versions to approve, failed rehearsals and skills whose software changed

The server's instructions carry the learning protocol: explore with a goal, note what you learn, finish, read the digest, save the skill; next time launch its environment and run its procedure.

Try it

"Spin up an ubuntu-22.04-headless capsule, run my test suite, and show me the output."

The model calls capsulate_launch, then capsulate_exec, and reports back.

"Launch an ubuntu-22.04-desktop capsule, give me a live link, and find the latest Firefox release notes. If anything needs a login, hand it to me."

The model launches the desktop, calls capsulate_console_link and gives you the URL, narrates with capsulate_say as it works, and calls capsulate_handoff when it needs you. You press Done, hand back in the live console when you finish, and it carries on.

Develop

bash
npm run build                          # from the repo root: sdk, ai-tools, cli, mcpnode apps/mcp/scripts/smoke-live.mjs   # live tools end to end against a fake API

来源：apps/mcp/README.md，提交 36f738e

工具

0
SourceWeft 尚未在沙箱中启动过该软件包，因此其工具未知。

版本历史

6
  1. v0.7.0最新Oct 13, 2026
  2. v0.6.0Oct 12, 2026
  3. v0.5.0Oct 11, 2026
  4. v0.4.1Oct 10, 2026
  5. v0.4.0Oct 10, 2026
  6. v0.2.0Sep 16, 2026

评分

已满足适用检查项的 50%

该评分衡量此收录条目的完整程度，依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价，也不表示服务器安全或当前可用：可连通性是带日期的检查结果，评分也不包含安装量或用户评价。

  • 近 30 天内可连通25 分

    请保持端点能够响应 MCP 握手；SourceWeft 每 30 天探测一次。

  • 已声明认证方式10 分

    请在注册中心条目中声明凭据请求头，或保持端点能够响应探测，以便判定其认证方式。

  • 已列出工具15 分

    请让服务器在无凭据时响应 tools/list，或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分