# a11ydriver-mcp

Source: https://glimind.com/server/npm%2Fa11ydriver-mcp
Registry id: `npm/a11ydriver-mcp`
Status: unknown

## Install

```json
{
  "mcpServers": {
    "a11ydriver-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "a11ydriver-mcp"
      ]
    }
  }
}
```

Claude Code:

```sh
claude mcp add a11ydriver-mcp -- npx -y a11ydriver-mcp
```

## Tools (26)

- `browser_cleanup_agent_group`: Close all tabs inside an agent tab group after task finishes
- `browser_click`: Click element by coordinates (x, y) or element refId
- `browser_close_tab`: Close specified tab by tabId
- `browser_create_agent_group`: Create an isolated tab group for AI agent session
- `browser_drag_and_drop`: Perform drag & drop from (fromX, fromY) to (toX, toY)
- `browser_emulate_device`: Emulate mobile screen viewport and device metrics
- `browser_evaluate_script`: Execute custom JS snippet on web context via CDP Runtime
- `browser_get_a11y_tree`: Get semantic Accessibility Tree snapshot with refId
- `browser_get_console_logs`: Retrieve JS console error/warning history
- `browser_get_cookies`: Get session cookies for active domain
- `browser_go_back`: Go back in browser history
- `browser_go_forward`: Go forward in browser history
- `browser_handle_dialog`: Accept or dismiss JS alert(), confirm(), prompt() popups
- `browser_hover`: Hover cursor over element or (x, y) coordinates
- `browser_launch`: Auto-launch Google Chrome with A11yDriver extension preloaded
- `browser_list_tabs`: List all open Chrome tabs with IDs & titles
- `browser_navigate`: Navigate active tab to target url
- `browser_new_tab`: Open a new tab (optional url, groupId)
- `browser_press_key`: Press special keys (Enter, Tab, Escape, Backspace, ArrowDown, ArrowUp)
- `browser_print_to_pdf`: Print web page to PDF base64 document
- `browser_reload`: Reload current page
- `browser_screenshot`: Capture tab screenshot (supports fullPage: true)
- `browser_scroll`: Scroll page vertically (deltaY) or horizontally (deltaX)
- `browser_switch_tab`: Switch active focused tab by tabId
- `browser_type`: Type text into element refId or active input
- `browser_upload_file`: Upload local file to file input refId

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.