# Browser automation tools for AI agents

> 4 browser automation listings ranked by the Anchor benchmark. Leader Chrome DevTools MCP (BB). Servers that let an agent drive a real browser. Navigate, read accessibility trees, click, fill forms, capture traces and screenshots.

- Canonical: https://www.anchorterminal.com/categories/browser
- Markdown: https://www.anchorterminal.com/categories/browser.md (~2,400 tokens)
- Slim: https://www.anchorterminal.com/categories/browser.min.md (~330 tokens, same facts, less prose, for token-sensitive contexts)
- JSON: https://www.anchorterminal.com/categories/browser.json (this page as data, same URL with Accept: application/json)
- Site index for agents: https://www.anchorterminal.com/llms.txt (full text: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- Updated: 2026-10-04

Servers that let an agent drive a real browser. Navigate, read accessibility trees, click, fill forms, capture traces and screenshots.

- Tools ranked: 4 · agent-ready (BB or better): 2 · accept x402: 1 · hosted endpoints: 1 · desk reviews by the panel: 20
- JSON: https://www.anchorterminal.com/api/v1/tools.json (list) · https://www.anchorterminal.com/api/v1/rankings.json (ranked) · https://www.anchorterminal.com/api/v1/x402.json (payable) · https://www.anchorterminal.com/api/v1/capabilities.json (by capability)
- Grades run AA, A, BB, B, C, D, E, F · methodology: https://www.anchorterminal.com/benchmark/

- Capabilities in this category: browser.control, browser.debug
- https://letme.dev/browser.control picks the top-graded tool in this list and says how to call it direct; calling through letme comes later (https://www.anchorterminal.com/letme/index.md)

## Ranking

| # | Tool | Vendor | Kind | Category | Grade | Score | Confidence | x402 | Auth | Where | Reviews | Page |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 22 | Chrome DevTools MCP | Google (Chrome DevTools team) | MCP server | Browser | BB | 77.1 | high | no | None | local | 3.5/5 (8) | https://www.anchorterminal.com/tools/chrome-devtools-mcp.md |
| 25 | Browserbase | Browserbase | HTTP API | Browser | BB | 76.6 | medium | yes | API key | hosted | 3.4/5 (8) | https://www.anchorterminal.com/tools/browserbase.md |
| 138 | Playwright MCP | Microsoft | MCP server | Browser | B | 67.6 | medium | no | None | local | 3/5 (2) | https://www.anchorterminal.com/tools/playwright-mcp.md |
| 443 | Puppeteer (archived MCP reference server) | Model Context Protocol (archived) | MCP server | Browser | F | 30.8 | high | no | None | local | 1/5 (2) | https://www.anchorterminal.com/tools/puppeteer-reference-server-archived.md |

Scores are from public evidence against the published checklist (https://www.anchorterminal.com/benchmark/), with Performance and Task success pending. p95 latency and context cost come from our probes, which haven't run yet.

## Summaries

### 22. Chrome DevTools MCP, BB (77.1)

Lets coding agents control and inspect a live Chrome instance. Performance traces, network inspection, heap snapshots and Lighthouse audits in one server. Usage statistics go to Google by default until you pass `--no-usage-statistics`.

- Page: https://www.anchorterminal.com/tools/chrome-devtools-mcp · Markdown: https://www.anchorterminal.com/tools/chrome-devtools-mcp.md · JSON: https://www.anchorterminal.com/api/v1/tools/chrome-devtools-mcp.json
- Capabilities: browser.control, browser.debug

### 25. Browserbase, BB (76.6)

Hosted headless browsers for agents over CDP, plus Fetch and Search APIs and the Stagehand framework. x402 sessions at x402.browserbase.com with no account, $0.12 an hour, unused minutes refunded. The hosted MCP setup page passes the API key as `?browserbaseApiKey=` in the URL.

- Page: https://www.anchorterminal.com/tools/browserbase · Markdown: https://www.anchorterminal.com/tools/browserbase.md · JSON: https://www.anchorterminal.com/api/v1/tools/browserbase.json
- Capabilities: browser.hosted, browser.control, web.fetch, web.search, web.scrape · endpoint: `https://api.browserbase.com/v1`

### 138. Playwright MCP, B (67.6)

Browser automation MCP server built on Playwright that drives pages via structured accessibility snapshots rather than screenshots. Accessibility snapshots with `depth` limits and `browser_find`, so the agent reads less of the page. `browser_run_code_unsafe` runs arbitrary JavaScript in the server process and can't be switched off.

- Page: https://www.anchorterminal.com/tools/playwright-mcp · Markdown: https://www.anchorterminal.com/tools/playwright-mcp.md · JSON: https://www.anchorterminal.com/api/v1/tools/playwright-mcp.json
- Capabilities: browser.control

### 443. Puppeteer (archived MCP reference server), F (30.8)

The original browser-automation reference server, driving Chrome through Puppeteer with screenshots and JavaScript evaluation. Seven tools in about 700 tokens, every input with a `required` list. Archived on 29 May 2025 with no security guarantees, and deprecated on npm.

- Page: https://www.anchorterminal.com/tools/puppeteer-reference-server-archived · Markdown: https://www.anchorterminal.com/tools/puppeteer-reference-server-archived.md · JSON: https://www.anchorterminal.com/api/v1/tools/puppeteer-reference-server-archived.json
- Capabilities: browser.control

## Indexed, not reviewed (21)

Sorted into this category from public catalogues, with facts and our own checks but no score, grade or rank (https://www.anchorterminal.com/indexed/index.md).

| Listing | Kind | What it does | Why it's here |
| --- | --- | --- | --- |
| [AgentUtility Browser Workflow](https://www.anchorterminal.com/tools/agentutility-mcp-browser-workflow.md) | MCP server | MCP server for the @agentutility browser-workflow cluster — pay-per-call x402 tools, no API keys… | vendor's own |
| [Aginx Browser](https://www.anchorterminal.com/tools/aginxbrowser.md) | MCP server | Stealth web browser for agents: search, fetch, click, download and type in persistent MCP sessions. | vendor's own |
| [AppContext](https://www.anchorterminal.com/tools/appcontext.md) | MCP server | Live iOS Simulator and browser screenshots, logs, and a personal MCP hub for AI coding agents. | vendor's own |
| [E2LLM](https://www.anchorterminal.com/tools/e2llm-element-to-llm.md) | MCP server | E2LLM gives your AI eyes and hands in a real browser: structured perception (SiFR) plus action. | vendor's own |
| [foura.ai MCP server](https://www.anchorterminal.com/tools/foura-mcp.md) | MCP server | Reliable web access for AI agents: smart HTTP, rotating proxies, and full-browser rendering. | vendor's own |
| [gamedev.pl](https://www.anchorterminal.com/tools/gamedev-creator.md) | MCP server | Build browser games on gamedev.pl from your coding agent. | vendor's own |
| [ghostlight](https://www.anchorterminal.com/tools/sylin-ghostlight.md) | MCP server | Visible local browser automation in signed-in Chromium, with optional policy and audit. | vendor's own |
| [maxoperf](https://www.anchorterminal.com/tools/maxoperf.md) | MCP server | Load & browser performance testing — drive MaxoPerf from your AI agent with your API key. | vendor's own |
| [openbrowser-ai](https://www.anchorterminal.com/tools/openbrowser-ai.md) | MCP server | AI browser automation. Write async Python to navigate, click, type, and extract data. | vendor's own |
| [PilotGentic](https://www.anchorterminal.com/tools/pilotgentic-mcp-server.md) | MCP server | macOS desktop control for Claude: click, type, screenshot, AX tree, browser control, agentic loops. | vendor's own |
| [Playrunner](https://www.anchorterminal.com/tools/playrunner-mcp.md) | MCP server | Discover Playwright workflows, start runs, and inspect results in Playrunner Cloud. | vendor's own |
| [Presio](https://www.anchorterminal.com/tools/presio.md) | MCP server | Present PDFs from the browser: upload to start a slideshow or validate Presio sidecars. No auth. | vendor's own |
| [QuickPeek](https://www.anchorterminal.com/tools/quickpeek.md) | MCP server | Turn a URL into a narrated demo video: real browser walkthrough, AI voiceover, captions burned in. | vendor's own |
| [reflex](https://www.anchorterminal.com/tools/reflexmcp-reflex.md) | MCP server | Browser automation for AI agents: compressed page snapshots, batched actions, fewer tokens. | vendor's own |
| [Screen Browser](https://www.anchorterminal.com/tools/screenbrowser.md) | MCP server | Narrated demo and tutorial videos of your web app, from a guide or your coding agent. | vendor's own |
| [Shiplight](https://www.anchorterminal.com/tools/shiplight.md) | MCP server | An MCP server that provides browser automation | vendor's own, widely used |
| [Tada: automated QA for AI-written code](https://www.anchorterminal.com/tools/taada-tada.md) | MCP server | Your coding agent asks Tada to work through a feature in a real browser and gets back a verdict. | vendor's own |
| [TurnSignal](https://www.anchorterminal.com/tools/turnsignal.md) | MCP server | Read-only Playwright CI results for agents: new vs known failures, flaky tests, test history. | vendor's own, widely used |
| [val](https://www.anchorterminal.com/tools/nyx-intelligence-val.md) | MCP server | QA agent for vibecoded apps. Real browser checks, your AI fixes them. Ship to green. | vendor's own |
| [walkerOS Source Browser](https://www.anchorterminal.com/tools/walkeros-source-browser.md) | MCP server | Generate, parse, and validate walkerOS data-elb HTML tagging attributes | vendor's own, widely used |
| [Web Browser (kogiQA)](https://www.anchorterminal.com/tools/atagon-kogiqa-mcp.md) | MCP server | This web browser has been designed to help your agent debug and develop complex web applications. | vendor's own |