A Model Context Protocol (MCP) server that provides browser automation capabilities using Playwright, forked to enhance anti-bot detection bypass. This server enables LLMs to interact with web pages through structured accessibility snapshots.

• Fast and lightweight: Uses Playwright's accessibility tree, not pixel-based input.
• LLM-friendly: No vision models needed, operates purely on structured data.
• Deterministic tool application: Avoids ambiguity common with screenshot-based approaches.

This project (playwright-mcp-bypass) is maintained by yan5xu and is available at yan5xu/playwright-mcp-bypass.

It originated as a fork of microsoft/playwright-mcp with the primary goal of enhancing the ability to bypass anti-bot detection mechanisms employed by some websites. This is achieved by adding the --disable-blink-features=AutomationControlled argument to the browser launch options, making the automated browser appear more like a regular user's browser.

• Web navigation and form-filling
• Data extraction from structured content
• Automated testing driven by LLMs
• General-purpose browser interaction for agents

{
  "mcpServers": {
    "playwright-mcp-bypass": {
      "command": "npx",
      "args": [
        "playwright-mcp-bypass@latest"
      ]
    }
  }
}

Install the Playwright MCP server in VS Code using one of these buttons:

Alternatively, you can install the Playwright MCP server using the VS Code CLI:

# For VS Code
code --add-mcp '{"name":"playwright-mcp-bypass","command":"npx","args":["playwright-mcp-bypass@latest"]}'

# For VS Code Insiders
code-insiders --add-mcp '{"name":"playwright-mcp-bypass","command":"npx","args":["playwright-mcp-bypass@latest"]}'

After installation, the Playwright MCP server will be available for use with your GitHub Copilot agent in VS Code.

The Playwright MCP server supports the following command-line options:

• --browser <browser>: Browser or chrome channel to use. Possible values:
  • chrome, firefox, webkit, msedge
  • Chrome channels: chrome-beta, chrome-canary, chrome-dev
  • Edge channels: msedge-beta, msedge-canary, msedge-dev
  • Default: chrome
• --caps <caps>: Comma-separated list of capabilities to enable, possible values: tabs, pdf, history, wait, files, install. Default is all.
• --cdp-endpoint <endpoint>: CDP endpoint to connect to
• --executable-path <path>: Path to the browser executable
• --headless: Run browser in headless mode (headed by default)
• --port <port>: Port to listen on for SSE transport
• --http-port <port>: Port to listen on for the request/response HTTP API.
• --user-data-dir <path>: Path to the user data directory
• --vision: Run server that uses screenshots (Aria snapshots are used by default)

Playwright MCP will launch the browser with the new profile, located at

- `%USERPROFILE%\AppData\Local\ms-playwright\mcp-chrome-profile` on Windows
- `~/Library/Caches/ms-playwright/mcp-chrome-profile` on macOS
- `~/.cache/ms-playwright/mcp-chrome-profile` on Linux

All the logged in information will be stored in that profile, you can delete it between sessions if you'd like to clear the offline state.

This mode is useful for background or batch operations.

{
  "mcpServers": {
    "playwright-mcp-bypass": {
      "command": "npx",
      "args": [
        "playwright-mcp-bypass@latest",
        "--headless"
      ]
    }
  }
}

When running headed browser on system w/o display or from worker processes of the IDEs, run the MCP server from environment with the DISPLAY and pass the --port flag to enable SSE transport.

npx playwright-mcp-bypass@latest --port 8931

And then in MCP client config, set the url to the SSE endpoint:

{
  "mcpServers": {
    "playwright-mcp-bypass": {
      "url": "http://localhost:8931/sse"
    }
  }
}
### HTTP API Usage (Request/Response)

In addition to the default Stdio transport and the SSE transport (`--port`), this server provides a standard HTTP API for request/response interactions. This is useful for clients that prefer simple HTTP calls over persistent connections.

#### Enabling the HTTP API

To enable the HTTP API, use the `--http-port` command-line option:

```bash
npx playwright-mcp-bypass@latest --http-port 8080

The server will then listen on the specified port (e.g., 8080) for incoming HTTP requests.

• Base Path: /tools/{tool_name}
• Method:
  • POST: Used for executing most tools. Tool parameters are sent in the JSON request body.
  • GET: Can be used for specific read-only tools like browser_tab_list. No request body is needed.
• Examples:
  • POST /tools/browser_navigate
  • POST /tools/browser_click
  • GET /tools/browser_tab_list

• Headers:
  • Content-Type: application/json
  • Session-Id: <your_session_id> (Optional, see Session Management)
• Body: A JSON object containing the parameters required by the specific tool.

Example (browser_navigate):

{
  "url": "https://example.com"
}

Example (browser_click):

{
  "element": "Login Button",
  "ref": "button#login"
}

• Content-Type: application/json
• Success (HTTP 200):

  {
    "success": true,
    "result": { ... } // The result returned by the tool execution
  }

• Error (HTTP 4xx/5xx):

  {
    "success": false,
    "error": "Error message describing the failure"
  }

The HTTP API manages browser state using sessions. Each session corresponds to an independent browser instance with its own context (unless a global --user-data-dir is specified).

• Session ID: Sessions are identified by the Session-Id HTTP header in the request.
• Default Session: If the Session-Id header is not provided, a default session named "default" is used.
• Session Creation: A new browser instance is automatically created when a request with a previously unseen Session-Id (or no ID for the default session) is received.
• Session Reuse: Subsequent requests with the same Session-Id will reuse the existing browser instance for that session.
• Session Timeout: Sessions automatically time out and close after 30 minutes of inactivity to conserve resources. Any request to an active session resets the timer.
• User Data Directory: By default, each session gets its own isolated user data directory (e.g., ~/.cache/ms-playwright/mcp-chromium-profile-<session_id>). If you specify --user-data-dir when starting the server, all HTTP sessions will share that single directory, which can lead to conflicts and is generally not recommended for concurrent sessions.

• SSE (--port): Establishes a persistent connection per client. State (browser instance) is tied to the connection lifetime. Communication is typically streaming (server sends events).
• HTTP API (--http-port): Uses standard request/response cycles. State is managed via the Session-Id header and has a timeout. Simpler for clients that don't need persistent connections.

# Navigate (uses default session if Session-Id header is omitted)
curl -X POST http://localhost:8080/tools/browser_navigate \
     -H "Content-Type: application/json" \
     -d '{ "url": "https://example.com" }'

# Click an element in a specific session
curl -X POST http://localhost:8080/tools/browser_click \
     -H "Content-Type: application/json" \
     -H "Session-Id: my-session-123" \
     -d '{ "element": "Login Button", "ref": "button#login" }'

# Get tab list (GET request, uses default session)
curl http://localhost:8080/tools/browser_tab_list

# Get tab list for a specific session
curl -H "Session-Id: my-session-123" http://localhost:8080/tools/browser_tab_list

// Navigate in default session
fetch('http://localhost:8080/tools/browser_navigate', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({ url: 'https://example.com' })
})
.then(response => response.json())
.then(data => console.log(data));

// Type text in a specific session
fetch('http://localhost:8080/tools/browser_type', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'Session-Id': 'user-abc-session'
  },
  body: JSON.stringify({
    element: "Search Input",
    ref: "input[name='q']",
    text: "Playwright MCP"
  })
})
.then(response => response.json())
.then(data => console.log(data));

// Get tab list for a specific session
fetch('http://localhost:8080/tools/browser_tab_list', {
  headers: {
    'Session-Id': 'user-abc-session'
  }
})
.then(response => response.json())
.then(data => console.log(data));

import requests
import json

base_url = 'http://localhost:8080/tools'
session_id = 'python-session-456'

# Navigate in a specific session
headers = {
    'Content-Type': 'application/json',
    'Session-Id': session_id
}
payload = {'url': 'https://github.com'}
response = requests.post(f'{base_url}/browser_navigate', headers=headers, data=json.dumps(payload))
print(response.json())

# Get tab list for the same session
headers_no_content = {'Session-Id': session_id}
response = requests.get(f'{base_url}/browser_tab_list', headers=headers_no_content)
print(response.json())

### Tool Modes

The tools are available in two modes:

1. **Snapshot Mode** (default): Uses accessibility snapshots for better performance and reliability
2. **Vision Mode**: Uses screenshots for visual-based interactions

To use Vision Mode, add the `--vision` flag when starting the server:

```js
{
  "mcpServers": {
    "playwright-mcp-bypass": {
      "command": "npx",
      "args": [
        "playwright-mcp-bypass@latest",
        "--vision"
      ]
    }
  }
}

Vision Mode works best with the computer use models that are able to interact with elements using X Y coordinate space, based on the provided screenshot.

import { createServer } from '@playwright/mcp';

// ...

const server = createServer({
  launchOptions: { headless: true }
});
transport = new SSEServerTransport("/messages", res);
server.connect(transport);

• browser_click

  • Description: Perform click on a web page
  • Parameters:
    • element (string): Human-readable element description used to obtain permission to interact with the element
    • ref (string): Exact target element reference from the page snapshot

• browser_hover

  • Description: Hover over element on page
  • Parameters:
    • element (string): Human-readable element description used to obtain permission to interact with the element
    • ref (string): Exact target element reference from the page snapshot

• browser_drag

  • Description: Perform drag and drop between two elements
  • Parameters:
    • startElement (string): Human-readable source element description used to obtain permission to interact with the element
    • startRef (string): Exact source element reference from the page snapshot
    • endElement (string): Human-readable target element description used to obtain permission to interact with the element
    • endRef (string): Exact target element reference from the page snapshot

• browser_type

  • Description: Type text into editable element
  • Parameters:
    • element (string): Human-readable element description used to obtain permission to interact with the element
    • ref (string): Exact target element reference from the page snapshot
    • text (string): Text to type into the element
    • submit (boolean, optional): Whether to submit entered text (press Enter after)
    • slowly (boolean, optional): Whether to type one character at a time. Useful for triggering key handlers in the page. By default entire text is filled in at once.

• browser_select_option

  • Description: Select an option in a dropdown
  • Parameters:
    • element (string): Human-readable element description used to obtain permission to interact with the element
    • ref (string): Exact target element reference from the page snapshot
    • values (array): Array of values to select in the dropdown. This can be a single value or multiple values.

• browser_snapshot

  • Description: Capture accessibility snapshot of the current page, this is better than screenshot
  • Parameters: None

• browser_take_screenshot

  • Description: Take a screenshot of the current page. You can't perform actions based on the screenshot, use browser_snapshot for actions.
  • Parameters:
    • raw (boolean, optional): Whether to return without compression (in PNG format). Default is false, which returns a JPEG image.

• browser_screen_move_mouse

  • Description: Move mouse to a given position
  • Parameters:
    • element (string): Human-readable element description used to obtain permission to interact with the element
    • x (number): X coordinate
    • y (number): Y coordinate

• browser_screen_capture

  • Description: Take a screenshot of the current page
  • Parameters: None

• browser_screen_click

  • Description: Click left mouse button
  • Parameters:
    • element (string): Human-readable element description used to obtain permission to interact with the element
    • x (number): X coordinate
    • y (number): Y coordinate

• browser_screen_drag

  • Description: Drag left mouse button
  • Parameters:
    • element (string): Human-readable element description used to obtain permission to interact with the element
    • startX (number): Start X coordinate
    • startY (number): Start Y coordinate
    • endX (number): End X coordinate
    • endY (number): End Y coordinate

• browser_screen_type

  • Description: Type text
  • Parameters:
    • text (string): Text to type
    • submit (boolean, optional): Whether to submit entered text (press Enter after)

• browser_press_key

  • Description: Press a key on the keyboard
  • Parameters:
    • key (string): Name of the key to press or a character to generate, such as ArrowLeft or a

• browser_tab_list

  • Description: List browser tabs
  • Parameters: None

• browser_tab_new

  • Description: Open a new tab
  • Parameters:
    • url (string, optional): The URL to navigate to in the new tab. If not provided, the new tab will be blank.

• browser_tab_select

  • Description: Select a tab by index
  • Parameters:
    • index (number): The index of the tab to select

• browser_tab_close

  • Description: Close a tab
  • Parameters:
    • index (number, optional): The index of the tab to close. Closes current tab if not provided.

• browser_navigate

  • Description: Navigate to a URL
  • Parameters:
    • url (string): The URL to navigate to

• browser_navigate_back

  • Description: Go back to the previous page
  • Parameters: None

• browser_navigate_forward

  • Description: Go forward to the next page
  • Parameters: None

• browser_press_key
  • Description: Press a key on the keyboard
  • Parameters:
    • key (string): Name of the key to press or a character to generate, such as ArrowLeft or a

• browser_file_upload

  • Description: Choose one or multiple files to upload
  • Parameters:
    • paths (array): The absolute paths to the files to upload. Can be a single file or multiple files.

• browser_pdf_save

  • Description: Save page as PDF
  • Parameters: None

• browser_wait

  • Description: Wait for a specified time in seconds
  • Parameters:
    • time (number): The time to wait in seconds (capped at 10 seconds)

• browser_close

  • Description: Close the page
  • Parameters: None

• browser_install

  • Description: Install the browser specified in the config. Call this if you get an error about the browser not being installed.
  • Parameters: None