The shortlist, before anything else: for daily tasks on your own accounts, ego (lite), which wins the login, token, and parallelism columns and costs nothing (full disclosure: we build it, and it's technically the list's one non-MCP entry). For debugging: Chrome DevTools MCP, unrivaled. For testing and CI: Playwright MCP.
What puts ego (lite) on top for daily work: it's an agent browser built for browser automation—Claude Code drives it through the ego-browser skill, writing whole workflows as scripts that execute outside the model's context, every site you've signed into stays signed in, and tasks run in parallel Spaces without touching your window.
If that's all you needed, you're done. The rest is the scoring method, all six options with their documented numbers, and the cases where the rankings flip, because a list that never flips is a list that wasn't scored honestly.
How is this list scored?
Four dimensions, weighted by how often they end up being the reason people switch tools: login state (can the agent act in your accounts without scripted auth), token cost (what a real multi-step task does to your context and bill), parallelism (can two tasks run without colliding, while you keep browsing), and price.
Maintenance is a pass/fail gate rather than a score: anything deprecated or abandoned is out. That's why no Puppeteer MCP appears here; the reference server is deprecated on npm.
On honesty: nobody has run all six through one identical task battery, so scores rest on each tool's published measurements and documented behavior, cited inline.
And one entry (ego (lite)) isn't an MCP server at all; it's an agent browser for browser automation that connects to Claude Code through the ego-browser skill. It's included because people searching "best browser MCP" want the capability (Claude Code driving a browser), not the protocol, and excluding the strongest answer on a technicality would make the list worse.
Capability first, protocol second.
The six options, one by one
1. ego (lite), agent browser
ego (lite) is an agent browser for browser automation: a free browser built for sharing your logged-in browser state with AI agents. Every site you've signed into stays signed in, and the agent inherits that state.
Claude Code drives it through the ego-browser skill, writing whole workflows as scripts that execute outside the model's context. In our published benchmark that meant 44% fewer execution rounds, 35.5% fewer tool calls, and 21.6% lower cost, and complex tasks run up to 3.45x faster than agent-browser.
Tasks run in parallel Spaces without touching your window. The limits, stated plainly: no debugging panels, no headless CI.
Setup is one command from the official repo, or one prompt to whatever agent you already run:
npx skills add citrolabs/ego-litePaste into your agent
Set up ego lite for me: https://github.com/citrolabs/ego-lite Read `skills/ego-browser/references/install.md` and follow the steps to install ego lite.
2. Playwright MCP
(Microsoft, free, Apache-2.0) is the default answer and the maintenance benchmark for this list: one-line setup, deterministic accessibility-tree refs, cross-browser, headless CI.
Its cost is structural, not configurable: measured test runs land at 89K-114K tokens, single complex-page snapshots pass 50K, and sessions degrade around step 12-15 on stale state.
Best testing tool here; worst daily-task economics.
3. Chrome DevTools MCP
(Google, free) owns the diagnostic kit nothing else on this list has: performance traces with Core Web Vitals, V8 heap snapshots, network forensics, device emulation.
Auto-connect (Chrome 144+) attaches it to your signed-in browser behind a permission dialog, at the price of sharing your window while it works. Install lean with --slim (3 tools) and enable the full kit per task.
What a snapshot actually costs here: a real take_snapshot call against Hacker News today, not paraphrased.
# take_snapshot via chrome-devtools-mcp@latest (mcp Python SDK, stdio session)
navigate_page chars: 123
take_snapshot chars: 38285
## Latest page snapshot
uid=1_0 RootWebArea "Hacker News" url="https://news.ycombinator.com/"
uid=1_1 link url="https://news.ycombinator.com/"
uid=1_2 link "Hacker News" url="https://news.ycombinator.com/news"
uid=1_3 StaticText "Hacker News"
uid=1_4 link "new" url="https://news.ycombinator.com/newest"
uid=1_5 StaticText "new"
uid=1_6 StaticText " | "
uid=1_7 link "past" url="https://news.ycombinator.com/front"
uid=1_8 StaticText "past"
uid=1_9 StaticText " | "
uid=1_10 link "comments" url="https://news.ycombinator.com/newcomments"
uid=1_11 StaticText "comments"
uid=1_12 StaticText " | "
uid=1_13 link "ask" url="https://news.ycombinator.com/ask"
...38,285 characters of accessibility tree for one page, roughly 9-10K tokens, before the agent has clicked anything. That's the substance behind this entry's token-cost row, and why DevTools MCP is the debugging pick rather than the daily-task pick.
4. Browser MCP
(browsermcp.io, free) is an extension-plus-server pair that connects Claude Code, Cursor, and other MCP clients to your existing browser: automation runs locally, "uses your existing browser profile, keeping you logged into all your services," and rides your real fingerprint.
The trade is the same as every extension route: it drives the browser you're using, one task at a time, in your window.
5. Browser Use (MCP mode)
(MIT, ~109K stars) is the autonomous framework speaking MCP: its navigation and extraction machinery exposed as tools for your client.
You inherit its strengths (resilience on unfamiliar pages) and its costs (model-heavy steps, and real-Chrome session attach documented as unreliable). Right when you specifically want its machinery inside an MCP workflow.
6. agent-browser (MCP mode)
(Vercel Labs, Apache-2.0, ~40K stars) is the Rust CLI's MCP face, with a core profile that keeps tool schemas small.
Fast and clean for stateless work. Sessions are isolated by design, so your logins stay out, and that's a feature until it's your blocker.
The scorecard
Read rows, not just the winner: the right pick flips completely depending on which column is your bottleneck, and every cell below traces to a documented measurement or an official doc named in this article.
| Option | Login state | Token cost | Parallelism | Price |
|---|---|---|---|---|
| Playwright MCP | None by default | Heaviest (measured) | Single sequential session by default | Free |
| Chrome DevTools MCP | Yes via auto-connect | Moderate; spills large bodies to disk | Your one window under auto-connect | Free |
| Browser MCP | Yes; your profile | Moderate; snapshot-based | Your one browser | Free |
| Browser Use MCP | Unreliable real-Chrome attach | Model-heavy per step | Multiple managed browsers | Free self-hosted |
| agent-browser MCP | None; isolated by design | Lean; filtered snapshots | Multiple isolated sessions | Free |
| ego (lite) | Yes; inherited | Leanest pattern (out-of-process, benchmarked) | Parallel Spaces, window untouched | Free |
Which should you pick, by what you do?
QA and test engineers: Playwright MCP for drafting and exploration, your existing framework for the suite, and budget the tokens (or use its CLI for long sessions). DevTools MCP joins the day a failure turns out to be performance-shaped.
Data collection and scraping: Public pages at volume: agent-browser or scripted Playwright. Sources behind your own logins: ego (lite), which starts past the sign-in and runs collections in parallel Spaces.
Daily assistant work (dashboards, forms, portals): ego (lite) first for the login state and the untouched window; Browser MCP as the pure-MCP alternative if you want extension-style simplicity and don't mind lending your browser while it runs.
One pattern across all three personas: the tool that wins your main workload rarely covers your edge cases, and that's fine. The MCP architecture (and ego (lite) as an agent browser alike) makes these cheap to keep side by side; the expensive mistake is forcing one tool to be all four columns.
Download ego (lite) for Mac, free, or see all five connection routes compared if you're still mapping the territory.
FAQ
What's the best browser MCP overall?
For strict MCP-only setups: Chrome DevTools MCP if your work includes any debugging, Playwright MCP if it's testing-centric, Browser MCP if login state matters most. Across capability regardless of protocol, ego (lite) as an agent browser wins the daily-task profile on logins, tokens, and parallelism together, free.
Is there a single MCP that does everything well?
No, and the scorecard shows why: for MCP servers the columns pull against each other. Login state wants your real browser; parallelism wants isolated browsers; debugging wants protocol depth. ego (lite) is the entry engineered to resolve that pull (inherited sessions inside isolated Spaces, execution outside the conversation), and its real concessions are debugging and headless CI, which is why the list still ends in pairings.
Can I install more than one?
Yes, and you should: they register independently and don't conflict. The practical pairing is one execution tool (ego (lite) or Playwright MCP by workload) plus DevTools MCP kept lean with --slim for diagnosis days. Watch total schema load: every enabled server's tool definitions sit in your context.
Why isn't there a Puppeteer MCP on the list?
The reference server-puppeteer is deprecated on npm with no maintained official successor, and a deprecated dependency holding a browser session fails this list's maintenance gate. Community forks exist; audit before trusting one.
How do MCP tool schemas affect my context budget?
Every enabled server's tool definitions load at session start: Playwright MCP's were measured at ~4,200 tokens across two dozen-plus tools, and stacking three browser servers triples that before any work happens. Keep one execution tool enabled by default, gate the rest behind per-task enabling, and use lean modes (--slim, --caps) where servers offer them.
Do these work with Cursor and Codex too?
All six: the five MCP options register through each client's MCP config, and ego (lite) attaches to any agent that can run a shell command, Cursor and Codex included. The scorecard transfers as-is.
