PluginBench
MCP Server
Maintained

io.github.sh6drack/zen-mcp MCP Server

io.github.sh6drack/zen-mcp

Automate Zen Browser from Claude and Cursor using WebDriver BiDi—no Selenium, no Playwright.

What is the io.github.sh6drack/zen-mcp MCP server?

The zen-mcp MCP server enables automation of Zen Browser through WebDriver BiDi protocol, allowing Claude Code and Cursor to control browser tabs, navigate pages, fill forms, take screenshots, and execute JavaScript. It connects directly via WebSocket without requiring external browser drivers like Selenium or Playwright.

zen-mcp provides 20 tools for browser automation: navigate to URLs, manage tabs, capture screenshots, extract page content, interact with forms and elements, and run JavaScript. Use it to automate web tasks, test web applications, scrape content, or control Zen Browser from AI agents in Claude Code or Cursor.

How to install io.github.sh6drack/zen-mcp

Copy-paste configuration for popular MCP clients.

transport: stdio
Config generated by PluginBench — verify against the source before use.
Environment / auth
  • ZEN_DEBUG_PORT

    Zen Browser remote debugging port

~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "zen-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "zen-mcp"
      ],
      "env": {
        "ZEN_DEBUG_PORT": "<YOUR_ZEN_DEBUG_PORT>"
      }
    }
  }
}

Tools & capabilities

Tools this server exposes to the agent.

  • zen_navigate — Go to a URL
  • zen_list_pages — List all open tabs
  • zen_select_page — Switch to a tab
  • zen_new_tab — Open a new tab
  • zen_close_tab — Close a tab
  • zen_snapshot — Page structure with selectors (filter: all/interactive/form)
  • zen_screenshot — Capture a screenshot
  • zen_get_page_text — Get page title, URL, and text
  • zen_get_form_fields — List all form fields with labels and values
  • zen_click — Click an element
  • zen_fill — Type into an input or textarea
  • zen_select_option — Pick a dropdown option
  • zen_check — Toggle a checkbox or radio
  • zen_press_key — Keyboard input (Enter, Tab, Ctrl+A, etc.)
  • zen_fill_form — Fill multiple fields at once
  • zen_scroll — Scroll the page or to an element
  • zen_evaluate — Run JavaScript in the page
  • zen_wait — Wait N milliseconds
  • zen_wait_for — Wait for text or element to appear
  • zen_reconnect — Force reconnect to Zen

Use cases

  • Automate web form filling and data entry tasks from Claude Code
  • Test web applications by navigating pages, clicking elements, and capturing screenshots
  • Extract page content and text for web scraping or analysis
  • Control Zen Browser tabs and navigation programmatically from Cursor
  • Run JavaScript in web pages to interact with React, Vue, and Angular applications

io.github.sh6drack/zen-mcp MCP server FAQ

What is zen-mcp?

zen-mcp is an MCP server that automates Zen Browser using WebDriver BiDi protocol. It provides 20 tools for navigation, form filling, screenshots, and JavaScript execution, accessible from Claude Code and Cursor without needing Selenium or Playwright.

Is zen-mcp free?

Yes, zen-mcp is open-source under the MIT license and free to use.

How do I install zen-mcp in Claude Code?

Install via npm with `npm install -g zen-mcp`, then add it to `~/.claude/mcp_servers.json` with the command `zen-mcp`. Start Zen with `--remote-debugging-port 9222` first.

How do I install zen-mcp in Cursor?

The installation process is the same as Claude Code: install via npm, add to your MCP servers configuration, and start Zen with remote debugging enabled.

What are the requirements?

You need Zen Browser installed and Node.js 20 or higher. Start Zen with the `--remote-debugging-port 9222` flag to enable WebDriver BiDi.

What are the known limitations?

File uploads and drag-and-drop are not supported due to WebDriver BiDi spec limitations. Some advanced BiDi commands available in Chrome may not work in Firefox's implementation. Zombie sessions require restarting Zen.

README (reference)

Source of truth, from the repository.

zen-mcp

The first MCP server for Zen Browser. Automate Zen from Claude Code, Cursor, or any MCP client.

No Selenium. No Playwright. No browser drivers. Just WebSocket.

Setup (2 minutes)

1. Start Zen with remote debugging

/Applications/Zen.app/Contents/MacOS/zen --remote-debugging-port 9222

Pro tip: Add alias zen='open /Applications/Zen.app --args --remote-debugging-port 9222' to your shell config. Then just run zen.

2. Add to Claude Code

# Option A: npm (recommended)
npm install -g zen-mcp

# Option B: Clone
git clone https://github.com/sh6drack/zen-mcp.git && cd zen-mcp && npm install

Add to ~/.claude/mcp_servers.json:

{
  "mcpServers": {
    "zen-browser": {
      "command": "zen-mcp"
    }
  }
}

If you cloned instead of npm install, use "command": "node", "args": ["/absolute/path/to/zen-mcp/server.mjs"]

Add to ~/.claude/settings.json:

{
  "permissions": {
    "allow": ["mcp__zen-browser__*"]
  }
}

That's it. Start a new Claude Code session and the zen_* tools are available.

20 Tools

Browse

ToolWhat it does
zen_navigateGo to a URL
zen_list_pagesList all open tabs
zen_select_pageSwitch to a tab
zen_new_tabOpen a new tab
zen_close_tabClose a tab

See

ToolWhat it does
zen_snapshotPage structure with selectors (filter: all/interactive/form)
zen_screenshotCapture a screenshot
zen_get_page_textGet page title, URL, and text
zen_get_form_fieldsList all form fields with labels and values

Interact

ToolWhat it does
zen_clickClick an element
zen_fillType into an input or textarea
zen_select_optionPick a dropdown option
zen_checkToggle a checkbox or radio
zen_press_keyKeyboard input (Enter, Tab, Ctrl+A, etc.)
zen_fill_formFill multiple fields at once
zen_scrollScroll the page or to an element

Utility

ToolWhat it does
zen_evaluateRun JavaScript in the page
zen_waitWait N milliseconds
zen_wait_forWait for text or element to appear
zen_reconnectForce reconnect to Zen

How It Works

Claude Code  ──stdio/MCP──>  zen-mcp  ──WebSocket/BiDi──>  Zen Browser

zen-mcp speaks WebDriver BiDi (W3C standard) directly over WebSocket. Form filling uses native value setters with input/change event dispatch so React, Vue, and Angular apps work correctly.

What Works Well

  • Navigation, clicking, form filling — rock solid, handles React/Vue/Angular
  • Screenshots and page reads — reliable content extraction
  • Tab management — open, close, switch between tabs
  • JavaScript evaluation — run any code in the page context
  • Keyboard input — shortcuts, Enter, Tab, modifier combos

Known Limitations

  • Zen inherits Firefox's WebDriver BiDi implementation, which is still maturing. Some advanced BiDi commands that work in Chrome may not be available yet.
  • Zombie sessions can only be cleared by restarting Zen (BiDi session.end is connection-scoped). zen-mcp detects this and tells you what to do.
  • No file upload or drag-and-drop support (BiDi spec limitation).

Built-in Reliability

  • Auto-reconnect with exponential backoff if WebSocket drops
  • Zombie session recovery when a previous client crashed
  • Connection retry (3 attempts with backoff)
  • Clean shutdown on SIGINT/SIGTERM to prevent orphaned sessions

Troubleshooting

ProblemFix
"Cannot connect to Zen Browser"Start Zen with --remote-debugging-port 9222
"Maximum number of active sessions"Restart Zen: killall zen && zen
Connection keeps droppingUse zen_reconnect to force a fresh connection

Config

Env VariableDefaultDescription
ZEN_DEBUG_PORT9222Zen's remote debugging port

Requirements

Test

node test-e2e.mjs   # 21 tests, needs Zen running

License

MIT

Related MCP servers

Search and discover local community listings, classifieds, services and events.

0
Python
View repository →

MCP server for PayTo agreements and payment initiation in Australia.

View repository →

ShaBaasPay MCP server for PayTo agreements and payment initiation in Australia.

Governed property-management MCP: portfolio, leasing, accounting, maintenance, OAuth 2.1.

View repository →

Persistent memory for AI coding tools. 11 tools for any MCP-compatible AI tool.

0
TypeScript
MIT
View repository →
CYCyberLens logo

CyberLens

Maintained

Security scanning for websites, public repositories, and Open CLAW skills.

0
TypeScript
MIT
View repository →