Browser Tools MCP — 사용 사례, 설치 & 라이브 데모

Name: Browser Tools MCP Server
Author: AgentDeskAI

왜 쓰나요

핵심 기능

Reads YOUR Chrome, not a headless one — logged-in auth, extensions, cookies
Console logs, network requests, selected-element inspection
Built-in audits: Lighthouse-style perf, a11y, SEO, Best Practices, Next.js
All data stays local — no cloud roundtrip

라이브 데모

실제 사용 모습

browser-tools.replay ▶ 준비됨

0/0

설치

클라이언트 선택

~/Library/Application Support/Claude/claude_desktop_config.json · Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "browser-tools": {
      "command": "npx",
      "args": [
        "-y",
        "@agentdeskai/browser-tools-mcp@latest"
      ]
    }
  }
}

Claude Desktop → Settings → Developer → Edit Config 열기. 저장 후 앱 재시작.

~/.cursor/mcp.json · .cursor/mcp.json

{
  "mcpServers": {
    "browser-tools": {
      "command": "npx",
      "args": [
        "-y",
        "@agentdeskai/browser-tools-mcp@latest"
      ]
    }
  }
}

Cursor는 Claude Desktop과 동일한 mcpServers 스키마 사용. 프로젝트 설정이 전역보다 우선.

VS Code → Cline → MCP Servers → Edit

{
  "mcpServers": {
    "browser-tools": {
      "command": "npx",
      "args": [
        "-y",
        "@agentdeskai/browser-tools-mcp@latest"
      ]
    }
  }
}

Cline 사이드바의 MCP Servers 아이콘 클릭 후 "Edit Configuration" 선택.

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "browser-tools": {
      "command": "npx",
      "args": [
        "-y",
        "@agentdeskai/browser-tools-mcp@latest"
      ]
    }
  }
}

Claude Desktop과 같은 형식. Windsurf 재시작 후 적용.

~/.continue/config.json

{
  "mcpServers": [
    {
      "name": "browser-tools",
      "command": "npx",
      "args": [
        "-y",
        "@agentdeskai/browser-tools-mcp@latest"
      ]
    }
  ]
}

Continue는 맵이 아닌 서버 오브젝트 배열 사용.

~/.config/zed/settings.json

{
  "context_servers": {
    "browser-tools": {
      "command": {
        "path": "npx",
        "args": [
          "-y",
          "@agentdeskai/browser-tools-mcp@latest"
        ]
      }
    }
  }
}

context_servers에 추가. 저장 시 Zed가 핫 리로드.

claude mcp add browser-tools -- npx -y @agentdeskai/browser-tools-mcp@latest

한 줄 명령. claude mcp list로 확인, claude mcp remove로 제거.

사용 사례

실전 활용법: Browser Tools

Debug a live front-end bug using real console + network context

👤 Frontend engineers ⏱ ~15 min intermediate

언제 쓸까: A bug only reproduces in your actual session (logged in, specific data). You want the agent to read the console and network tab.

사전 조건

Chrome extension installed + bridge server running — Install extension from releases page, then npx @agentdeskai/browser-tools-server@latest in a terminal
MCP server wired to your client — npx -y @agentdeskai/browser-tools-mcp@latest

흐름

Reproduce the bug with DevTools open

Open the affected page in Chrome with the extension active. Trigger the bug. Don't close the tab.✓ 복사됨

→ Bug visible, extension icon shows it's capturing
Pull context into the agent

Get the last 50 console messages and any failed network requests. Summarize what the page is yelling about.✓ 복사됨

→ Specific errors surfaced with stack traces
Diagnose and propose fix

Based on the errors, point at the file/line likely responsible. If 404s, identify the bad URL; if JS errors, the root cause.✓ 복사됨

→ Concrete hypothesis with evidence

결과: A debugging loop where the agent has the same context as you opening DevTools — faster than reading a screenshot description.

함정

Bridge server not running, MCP silently returns empty data — Always check curl localhost:3025/identity responds first; the bridge must be running before the MCP call
Capturing from the wrong tab — Extension captures the active tab. Click into the right tab before asking the agent

함께 쓰기: sentry

Run a Lighthouse-style audit on the page you're looking at

👤 Frontend / SEO / a11y engineers ⏱ ~10 min beginner

언제 쓸까: You want to audit a page as it actually renders in your session (not a clean crawler).

흐름

Open the target page

Navigate to <URL> in Chrome with the extension active.✓ 복사됨

→ Page loaded
Run the audit mode

Run the audit tool — perf, a11y, SEO, best practices. Summarize top 5 issues with fix suggestions.✓ 복사됨

→ Prioritized issue list
Focus in

For the accessibility failures, list exact selectors and the WCAG rule violated.✓ 복사됨

→ Actionable remediation steps

결과: A full audit without leaving your browser, scoped to real user state.

함정

Audits for auth-gated pages fail because the tool re-navigates — Use the debugger mode's element/console tools on the already-loaded page instead of the full audit

Inspect and explain a specific element

👤 Frontend devs, designers ⏱ ~5 min beginner

언제 쓸까: Something looks off in one component. You want a quick 'what CSS is applied and why' without digging.

흐름

Select the element in DevTools

Open DevTools, click the element inspector, select the offending element.✓ 복사됨

→ Element highlighted in Elements panel
Ask the agent to inspect

Get the selected element's computed styles, inherited rules, and bounding box. Explain why it's rendering the way it is.✓ 복사됨

→ Explanation grounded in actual CSS cascade
Propose a fix

Suggest the smallest CSS change to achieve [desired visual]. Show the before/after rule.✓ 복사됨

→ Minimal diff suggestion

결과: CSS debugging at conversation speed.

함정

Element selection doesn't persist — agent sees a different one — Re-select the element right before asking; don't let the DevTools panel change focus between messages

조합

다른 MCP와 조합해 10배 효율

browser-tools + sentry

Reproduce a Sentry error in YOUR browser and pull live console context

Sentry issue WEB-3a91 says an XHR to /api/checkout returns 500. Navigate to the page in my Chrome, reproduce, and capture the console + network.✓ 복사됨

browser-tools + playwright

browser-tools for live debugging, Playwright for repeatable reproductions afterward

I found the bug in my live Chrome via browser-tools. Now write a Playwright test that reproduces the exact click sequence.✓ 복사됨

도구

이 MCP가 노출하는 것

도구	언제 호출	비용
takeScreenshot	Visual capture of current state	free
getConsoleLogs / getConsoleErrors	Debug any JS error	free
getNetworkLogs / getNetworkErrors	Diagnose failed API calls	free
getSelectedElement	User has selected an element in DevTools	free
runAccessibilityAudit	A11y review of current page	free
runPerformanceAudit	Perf audit of current page	free
runSEOAudit	SEO review	free
runBestPracticesAudit	General code quality review	free
runNextJSAudit	You're on Next.js	free
wipeLogs	Clear between investigation sessions	free

비용 및 제한

운영 비용

API 쿼터: None — all local
호출당 토큰: Log dumps can be big (5k+ tokens). Use wipeLogs between sessions.
금액: Free, open source
팁: Wipe logs before each investigation — avoids pulling in unrelated noise

보안

권한, 시크릿, 파급범위

자격 증명 저장: None — runs against your existing Chrome session

데이터 외부 송신: Only to localhost:3025 (the bridge). Nothing leaves your machine except via whatever the agent/MCP client does.

Extension has access to your browser data by design. Only install from official releases. Review source before granting permissions.

문제 해결

자주 발생하는 오류와 해결

All tools return empty / 'not connected'

The bridge server isn't running. Start it: npx @agentdeskai/browser-tools-server@latest in a dedicated terminal.

확인: curl http://localhost:3025/identity

Extension icon grey / inactive

Click the extension icon, confirm it says 'Connected'. If not, reload the tab.

Audits fail with timeout

Some audits need extra time on heavy pages. Close other tabs to free CPU; re-run.

Wrong tab captured

The extension captures the tab that was last activated. Click into the right tab immediately before calling a tool.

대안

Browser Tools 다른 것과 비교

대안	언제 쓰나	단점/장점
chrome-devtools MCP	You want direct CDP access without a browser extension	No extension needed but you lose real-user-session context
Playwright MCP	You need repeatable automation, not live debugging	Clean browser — can't see your logged-in session

Browser Tools