Browser automation
Published entries filed under “Browser automation” in MCP, newest first by publication date on this site.
5 entries
- Browser automation
Browser Control MCP: let an agent tidy your own Firefox
An MCP server plus Firefox extension: agents manage tabs, search history and, with consent, read page text. No clicks or scripts, so it suits a daily browser.
- Browser automation
Firefox DevTools MCP: drive and debug Firefox from an agent
Mozilla's official Firefox MCP server over WebDriver BiDi. Agents click and fill by snapshot, and read network requests, console logs and profiler data.
- Browser automation
Lightpanda: a lightweight headless browser for agents
A headless browser written from scratch in Zig with a built-in MCP server. Agents read pages as Markdown, click and fill forms without spinning up Chromium.
- Browser automation
Playwright MCP: browser control through page structure
Microsoft's official browser MCP server: agents act on the page's accessibility tree instead of screenshots, so a plain text model can drive the web reliably.
- Browser automation
WebdriverIO MCP: browsers, Electron and mobile apps in one server
WebdriverIO's official MCP server: agents drive Chrome, Firefox and Safari, plus native iOS and Android apps through Appium and Electron desktop apps.