For browser debugging, Selenium 1 and Playwright 4 are the established browser automation tools that drive real browsers for scripted inspection and testing, and Browserbase 3 offers trace capture of browser automation (browser-trace action 13) for debugging agent runs. Browser MCP 9 and Rove 10 let agents use a real browser that a human can watch and take over.
- 1
Playwrightplaywright.devwebsite read, tools not yet checked
Playwright provides auto-waiting, tracing and cross-browser automation for Chromium, Firefox and WebKit, suited to reproducing and debugging page behaviour.
official:SDK · playwright+230 community tools built for it
- 2
- 3
Browserbasebrowserbase.comverified and signed
Browserbase's browser-trace action captures a full DevTools-protocol trace of a browser automation run, including CDP events, screenshots and DOM dumps, for debugging.
official:MCP server · @browserbasehq/mcpskill · browserskill · agent-experienceskill · autobrowseskill · browser-to-api+9 community tools built for it
for agents: browser-trace · browser-to-api · browser
- 4
- 5
Roveroveapi.comverified and signed
Rove gives an agent a real browser streamed live, letting you watch the session and take control at the point of failure.
for agents: evaluate · list_sessions · navigate
Sources
- 1Seleniumselenium.dev · read 2026-10-29
- 3Browserbasebrowserbase.com · read 2026-10-29
- 4Fast and reliable end-to-end testing for modern web apps | Playwrightplaywright.dev · read 2026-10-29
- 9Browser MCP - Let AI use your real browserbrowsermcp.io · read 2026-10-29
- 10Rove — the supervised browser for agentsroveapi.com · read 2026-10-29
- 13browser-trace: declares "browser-trace" (the record we read and signed)unfragile.ai · read 2026-10-24
For agents: the 10 actions behind this answer
browser · WebDriver Support verified and signed
web-browser · web-browser verified and signed
Automate and interact with web pages through Chrome or Chromium using the Chrome DevTools Protocol (CDP): navigate, click, fill forms, inspect content, take screenshots, and debug console or network activity. Use when an agent needs a real browser. Prefer headless Chrome unless visible browser inter
browser-trace · browser-trace verified and signed
Capture a full DevTools-protocol trace of any browser automation — CDP firehose, screenshots, and DOM dumps — then bisect the stream into per-page searchable buckets. Use when the user wants to debug a failed run, audit network/console/DOM activity, attach a trace to an in-progress session, or feed
· Playwright website read, tools not yet checked
Web automation and testing for apps, scripts, and AI agents
browsercat_evaluate · BrowserCat MCP Server (Browser Automation) found in code, not yet checked
Execute JavaScript in the browser console
browser-harness · browser-harness verified and signed
Control a real browser via CDP: clicking, typing, navigation, logged-in sessions, JS-rendered or bot-protected pages. Not for plain HTTP fetches of public content - use curl for those.
playwright-cli · playwright-cli verified and signed
Automate browser interactions, test web pages and work with Playwright tests.
browser · browser verified and signed
Direct browser control via CDP. Use when the user wants to automate, scrape, test, or interact with web pages. Connects to the user's already-running Chrome.
browserless_sessions · browserless-mcp found in code, not yet checked
Inspect the sessions on the Browserless account behind the current API
· Browser MCP website read, tools not yet checked
Let Claude, Cursor, Codex and other AI apps use your real browser, with your logins, in tabs you can watch and stop.
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.