Browser Automation MCP

Example Use Cases

• 1

  Programmatically navigate web pages, interact with elements (click, fill, press keys), and perform end-to-end browser workflows using Playwright via MCP.

• 2

  Inspect and extract hierarchical DOM structure or specific element subtrees for scraping, validation, or selector generation.

• 3

  Execute arbitrary JavaScript in page context and retrieve telemetry such as console logs and network requests for debugging or monitoring automated runs.

Description

browser-mcp provides a Model Context Protocol (MCP) server backed by Playwright to navigate pages, inspect and interact with the DOM, execute page JavaScript, and collect console logs and network requests. It is designed for reliable, scriptable browser automation in both local development and production (Docker) environments. The server exposes a suite of tools (navigate, click, fill, execute_js, explore DOM, retrieve logs/requests, etc.) to enable agents to automate and verify web interactions robustly.