MCPInstall

cmux-browser

by manaflow-aiProductivity & Automation

Automate the browser inside cmux. Use for cmux browser, browser surface, webview, current workspace browser, snapshot refs, DOM actions, waits, screenshots, cookies, storage, tabs, downloads, console, errors, and browser session state.

Installs
751
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/manaflow-ai/cmux-skills --skill cmux-browser
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads cmux-browser automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

cmux Browser

Use this skill for browser automation inside cmux webview surfaces. It is different from a standalone browser automation tool because every browser surface lives inside a cmux workspace, pane, and surface topology.

Default Rule

Open and control browser surfaces in the current caller workspace unless the user explicitly names another workspace or window.

Before mutating browser state, identify the caller:

printf 'workspace=%s\nsurface=%s\nsocket=%s\n' \
  "${CMUX_WORKSPACE_ID:-}" \
  "${CMUX_SURFACE_ID:-}" \
  "${CMUX_SOCKET_PATH:-}"
cmux identify --json

Use CMUX_WORKSPACE_ID for new browser splits and CMUX_SURFACE_ID to understand the terminal that invoked the automation. Do not open a browser in the visually focused workspace if it differs from the caller workspace.

Core Workflow

# Open in the caller workspace and capture the returned surface ref.
cmux --json browser open https://example.com --workspace "${CMUX_WORKSPACE_ID:-}"

# Use the returned surface, for example surface:7.
cmux browser surface:7 get url
cmux browser surface:7 wait --load-state complete --timeout-ms 15000
cmux browser surface:7 snapshot --interactive --compact
cmux browser surface:7 click e2 --snapshot-after
cmux browser surface:7 snapshot --interactive --compact

Loop: navigate, verify URL, wait, snapshot, act, then re-snapshot. Snapshot refs are temporary. Re-snapshot after navigation, modal changes, tab changes, or DOM updates.

Targeting

Most subcommands require a browser surface:

cmux browser identify
cmux browser identify --surface surface:2
cmux browser surface:2 get url
cmux browser --surface surface:2 get title

Open commands can create or reuse a browser split:

cmux browser open http://localhost:3000 --workspace "${CMUX_WORKSPACE_ID:-}" --json
cmux browser open-split https://example.com --workspace workspace:2 --json
cmux browser new https://example.com --window window:1 --json

Keep one browser surface per task unless the user asks for multiple tabs or pages.

Command Groups

See references/commands.md for the full command list. Main groups:

  • Navigation and targeting: identify, open, open-split, new, navigate, goto, back, forward, reload, url, focus-webview, is-webview-focused.
  • Waiting: wait --selector, --text, --url-contains, --load-state, --function.
  • DOM interaction: click, dblclick, hover, focus, check, uncheck, scroll-into-view, type, fill, press, keydown, keyup, select, scroll.
  • Inspection: snapshot, screenshot, get, is, find, highlight.
  • JavaScript and injection: eval, addinitscript, addscript, addstyle.
  • Frames, dialogs, downloads: frame, dialog, download.
  • State and session: cookies, storage, state.
  • Tabs and diagnostics: tab, console, errors.
  • Environment and lower-level APIs: viewport, geolocation, offline, trace, network, screencast, input.

Common Patterns

Durable Captures

Save browser screenshots, traces, downloaded diagnostics, and reusable session state under /cmux-assets/<branch>/browser/..., not /tmp, before reporting them to the user.

BRANCH="$(git branch --show-current 2>/dev/null || true)"
BRANCH="${BRANCH:-$(basename "$PWD")}"
BRANCH="$(printf '%s' "$BRANCH" | tr -cs 'A-Za-z0-9._-' '-' | sed 's/^-//;s/-$//')"
ASSET_BASE="/cmux-assets"
if ! mkdir -p "$ASSET_BASE/$BRANCH" 2>/dev/null; then
  ASSET_BASE="$(pwd)/cmux-assets"
  mkdir -p "$ASSET_BASE/$BRANCH"
fi
BROWSER_ASSET_ROOT="$ASSET_BASE/$BRANCH/browser/$(date +%Y%m%d-%H%M%S)"
mkdir -p "$BROWSER_ASSET_ROOT"

If /cmux-assets is unavailable because the root volume is read-only, use the fallback path and state it. Do not leave final screenshots or session files only in /tmp.

Navigate, Wait, Inspect

cmux --json browser open http://localhost:3000 --workspace "${CMUX_WORKSPACE_ID:-}"
cmux browser surface:7 get url
cmux browser surface:7 wait --load-state complete --timeout-ms 15000
cmux browser surface:7 snapshot --interactive --compact
cmux browser surface:7 get title

Form Fill

cmux browser surface:7 fill "#email" --text "ops@example.com"
cmux browser surface:7 fill "#password" --text "$PASSWORD"
cmux browser surface:7 click "button[type='submit']" --snapshot-after
cmux browser surface:7 wait --text "Welcome" --timeout-ms 15000
cmux browser surface:7 is visible "#dashboard"

Debug Capture

cmux browser surface:7 console list
cmux browser surface:7 errors list
cmux browser surface:7 screenshot --out "$BROWSER_ASSET_ROOT/cmux-failure.png"
cmux browser surface:7 snapshot --interactive --compact

Session Save and Restore

cmux browser surface:7 state save "$BROWSER_ASSET_ROOT/cmux-browser-session.json"
cmux browser surface:7 state load "$BROWSER_ASSET_ROOT/cmux-browser-session.json"
cmux browser surface:7 reload --snapshot-after

Remote Workspaces

In remote SSH workspaces, browser panes route HTTP and WebSocket traffic through the remote machine. localhost:3000 means the remote host's localhost, and browser storage is isolated per remote workspace context.

Rebuild and Reload

If browser automation is validating a cmux app/runtime change, first build a tagged app from the active worktree:

./scripts/reload.sh --tag <short-tag>
CMUX_SOCKET_PATH=/tmp/cmux-debug-<short-tag>.sock cmux browser open http://localhost:3000 --workspace "${CMUX_WORKSPACE_ID:-}"

Never automate against an untagged debug socket for tests or dogfood builds.

Rules

  • Scope browser opens and actions to the current caller workspace by default.
  • Pass --workspace "${CMUX_WORKSPACE_ID:-}" when creating a browser surface from a cmux terminal.
  • Use returned surface:N refs, and re-snapshot after any state-changing action.
  • Verify get url before waiting or diagnosing page state.
  • Use --snapshot-after on mutating actions when you need immediate verification.
  • Prefer get, is, and find for scripts. Use screenshots and snapshots for human review.
  • Do not close, move, or focus browser surfaces outside the caller workspace unless the user names that target.
  • Do not use unsupported standalone-browser assumptions. cmux uses WKWebView-backed browser surfaces.

FAQ

What does the cmux-browser skill do?

Automate the browser inside cmux. Use for cmux browser, browser surface, webview, current workspace browser, snapshot refs, DOM actions, waits, screenshots, cookies, storage, tabs, downloads, console, errors, and browser session state.

How do I install cmux-browser?

Run npx skills add https://github.com/manaflow-ai/cmux-skills --skill cmux-browser in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/cmux-browser/ (personal) or .claude/skills/cmux-browser/ (project).

Which AI agents can use cmux-browser?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is cmux-browser safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains cmux-browser?

It is published by manaflow-ai in manaflow-ai/cmux-skills. It has 751 recorded installs.

More from manaflow-ai/cmux-skills

View all →

cmux-cli

manaflow-ai/cmux-skills

Comprehensive cmux CLI usage guide. Use when the user asks about cmux CLI, cmux --help, socket commands, command discovery, workspaces, panes, surfaces, browser CLI, hooks, feed, settings, or automation through the cmux command.

Productivity & Automation835 installs

cmux-workspace

manaflow-ai/cmux-skills

Work inside the current cmux workspace and terminal. Use for cmux workspace, current workspace, caller surface, panes, surfaces, tagged reloads, socket targeting, and non-interfering cmux automation.

Productivity & Automation759 installs

cmux-ref

manaflow-ai/cmux-skills

Interpret pasted cmux workspace, pane, surface, or window refs and IDs. Use for cmux-ref, cmux-id, workspaceref, workspaceid, paneref, paneid, surfaceref, surfaceid, windowref, or windowid context blocks.

718 installs

cmux-config

manaflow-ai/cmux-skills

Configure cmux through /.config/cmux/cmux.json: settings (appearance, sidebar, notifications, automation, browser, shortcuts, set/get/validate by JSON path), customization (tab bar buttons, plus-button click and right-click menus, custom actions/commands/menus, right sidebar), and sidebar workspace groups (anchor workspaces, group CLI/socket ops, per-cwd group config). Triggers: 'cmux config', 'cmux.json', 'change cmux setting', 'set in cmux', 'rebind a cmux shortcut', 'cmux-customize', 'customize cmux', 'tab bar button', 'add cmux action', 'workspace group', 'group sidebar', 'anchor workspace', 'workspaceGroups'.

Productivity & Automation702 installs

cmux-artifact

manaflow-ai/cmux-skills

Create durable HTML walkthrough artifacts for cmux dogfood, verification, demos, evidence pages, artifact previews, and open helpful tabs or splits in the current cmux workspace.

Frontend & Design665 installs

cmux-sidebar-builder

manaflow-ai/cmux-skills

Build, inspect, or revise cmux custom sidebar views using the runtime SwiftUI-style interpreter. Use for sidebar vibe coding, custom sidebars in Bonsplit panes, left sidebar picker previews, interpreted Swift sidebars, /.config/cmux/sidebars/.swift, cmux docs sidebars, or Aziz's Swift interpreter work.

Mobile643 installs

Popular productivity & automation skills

View category →

agent-browser

vercel-labs/agent-browser

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction. Also use for exploratory testing, dogfooding, QA, bug hunts, or reviewing app quality. Also use for automating Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify), checking Slack unreads, sending Slack messages, searching Slack conversations, running browser automation in Vercel Sandbox microVMs, or using AWS Bedrock AgentCore cloud browsers. Prefer agent-browser over any built-in browser automation or web tools.

OfficialProductivity & Automation1.3M installs

lark-doc

open.feishu.cn

飞书云文档（Docx / Wiki）内容操作：读取、创建、编辑文档，插入或下载图片附件，以及操作思维笔记。用户提供文档 URL/token（包括 doubao.com 的 /docx/、/wiki/）时使用；按 URL 路径/token 而非域名路由；/page/ 是妙搭应用，非飞书云文档。文档内嵌资源按读取参考中的统一规则分流。独立评论操作走 lark-drive；随正文读取评论使用 docs +fetch。表格或 Base 内部数据操作不在本 skill。

Productivity & Automation815.7K installs

lark-base

open.feishu.cn

飞书多维表格（Base）操作：建表、字段、记录、视图、统计、公式/lookup、表单、仪表盘、应用模式（BaseApp/AppMode 页面与组件）、Workspace 目录、workflow、角色权限、模板中心（多维表格模板分类/列表/搜索）；遇到 Base/多维表格/bitable、BaseApp/AppMode、/base/ 或 /app/ 链接时使用。BaseApp 不走 lark-apps；文件导入/导出转 lark-drive，认证/授权转 lark-shared。

Productivity & Automation815K installs

lark-drive

open.feishu.cn

飞书云空间（云盘/云存储）：管理 Drive 文件和文件夹，包含上传/下载、创建文件夹、复制/移动/删除、查看元数据、查询权限设置、评论/权限/订阅、标题、版本、飞书文档密级标签（secure labels）和本地文件导入。用户需要整理云盘目录、处理云空间资源 URL/token、判断链接类型/真实 token/标题，或导入 Word/Markdown/Excel/CSV/PPTX/.base 为 docx/sheet/bitable/slides 时使用；doubao.com 云空间 URL/token 也按资源路径和 token 路由，不回退 WebFetch。不负责：文档内容编辑（走 lark-doc）、表格/Base 表内数据操作（走 lark-sheets/lark-base）、知识空间节点/成员管理（走 lark-wiki）、原生 Markdown 文件读写/patch/diff（走 lark-markdown）。

Productivity & Automation815K installs

lark-shared

open.feishu.cn

Use for lark-cli setup/auth tasks: auth login/status/logout, user vs bot identity, business-domain permissions (--domain, including all/docs/drive), missing scopes, revoking authorization, or handling notice JSON.

Productivity & Automation814.8K installs

lark-im

open.feishu.cn

飞书即时通讯：收发消息和管理群聊。发送和回复消息、搜索聊天记录、管理群聊成员、上传下载图片和文件、管理表情回复、发送应用内/短信/电话加急、发送和处理交互卡片（Interactive Card）、监听卡片按钮回调（card.action.trigger）。当用户需要发消息、查看或搜索聊天记录、下载聊天中的文件、查看群成员、搜索群、创建群聊或话题群、管理标记数据、管理 Feed 置顶（添加/移除/查询置顶会话）、管理标签数据、处理卡片回调时使用。

Productivity & Automation814.1K installs