Skip to content
gentic.news — AI News Intelligence Platform
Connecting to the Living Graph…

Listen to today's AI briefing

Daily podcast — 5 min, AI-narrated summary of top stories

Developer using Claude Code to control a real Chrome browser with logged-in sessions via the Browser Bridge MCP…
Open SourceScore: 75

Browser Bridge MCP: Drive Your Real Logged-In Chrome from Claude Code

Browser Bridge MCP gives Claude Code control of your real Chrome session with 63 tools. Install in 60 seconds, then automate authenticated browsing, network capture, and security testing.

·2d ago·4 min read··21 views·AI-Generated·Report error
Share:
Source: github.comvia hn_claude_code, devto_claudecodeWidely Reported
How do I use Browser Bridge MCP to let Claude Code control my real Chrome session?

Browser Bridge is an MCP server + Chrome extension that lets Claude Code control your real, logged-in browser. Install it in 60 seconds, then ask Claude to read feeds, capture network traffic, or run security tests against live authenticated apps.

TL;DR

Browser Bridge lets Claude Code control your real Chrome session with cookies, SSO, and 2FA intact via 63 tools.

Key Takeaways

  • Browser Bridge MCP gives Claude Code control of your real Chrome session with 63 tools.
  • Install in 60 seconds, then automate authenticated browsing, network capture, and security testing.

What Changed — Control Your Real Chrome from Claude Code

I Tested (New) Claude Code Browser Feature (Claude Code Can ...

Browser Bridge is a new open-source MCP server + Chrome extension that lets Claude Code and OpenAI Codex CLI drive your actual logged-in Chrome browser. No headless browser. No fresh profile. No re-login. Your agent inherits your cookies, HttpOnly sessions, SSO, and 2FA state automatically.

It ships 63 tools covering:

  • Everyday browsing (navigate, click, extract text)
  • DevTools-grade network capture (see API traffic in real time)
  • Full web-security testing toolkit (inspect cookies, headers, XSS detection)
  • Playbooks (saved, self-healing recipes for repetitive tasks)
  • Session recording with replay (export to offline HTML or high-res MP4)

What It Means For You — Concrete Impact on Daily Claude Code Usage

Before Browser Bridge, if you wanted Claude Code to check a dashboard, read a feed, or test a feature behind authentication, you had two bad options:

  1. Use a headless browser with mocked cookies (brittle, often fails)
  2. Manually copy data and paste it into Claude (slow, breaks flow)

Now Claude Code can:

  • "Read my feed and summarize it" — works against your real, logged-in session
  • "Capture the API traffic on this page and show me the responses" — intercepts network requests via DevTools
  • "Run a security audit on this app" — uses the built-in security toolkit
  • "Record this interaction and export it as a video" — for bug reports or demos

Because the extension runs inside your real Chrome profile, it bypasses all the friction of headless browsers. No cookie extraction, no session token management, no 2FA workarounds.

Try It Now — Setup in 60 Seconds

1. Clone and build

git clone https://github.com/vitalysim/browser-bridge.git && cd browser-bridge
( cd server && npm install && npm run build )
( cd extension && npm install && npm run build )
( cd server && npm run install-service )  # autostart at login

2. Load the extension

  • Open chrome://extensions
  • Enable Developer mode
  • Click "Load unpacked" and select the extension/ folder
  • Open its options and paste the token from ~/.browser-bridge/token
  • Click Save & connect

3. Connect Claude Code

claude mcp add --transport http --scope user browser-bridge \
  http://127.0.0.1:8765/mcp \
  --header "Authorization: Bearer $(cat ~/.browser-bridge/token)"

4. Verify

curl -s http://127.0.0.1:8765/health
# → {"ok":true,"extensionConnected":true}

5. Use it

claude "Navigate to my Jira dashboard and summarize the open tickets"
claude "Open the admin panel and capture all API traffic"
claude "Record a session of me filing a bug report and export it as MP4"

Architecture — Why This Works Better

Browser Bridge uses a local MCP server + Chrome extension approach instead of Chromium's remote debug protocol. This means:

  • The agent inherits your real profile's cookies, HttpOnly sessions, SSO, and 2FA state
  • No need to extract tokens or manage session state
  • One server can serve both Claude Code and Codex CLI simultaneously

Limitations & Security

  • Chrome only (Manifest V3 extension). No Firefox/Safari support yet.
  • Local only — the server binds to 127.0.0.1:8765 and uses token authentication
  • Your agent has full browser access — only connect trusted AI agents
  • Not for production automation — designed for development, testing, and personal use

The roadmap includes Firefox support, multi-tab orchestration, and deeper DevTools integration.

Why This Matters for Claude Code Users

Browser Bridge fills a critical gap in the MCP ecosystem. While tools like k-dense ship scientific agent skills, Browser Bridge gives Claude Code real browser access. This is especially powerful combined with Claude Opus 5's 0% prompt injection rate in browser agents — you can trust Claude to safely interact with authenticated sessions.

For developers building web apps, this means: ask Claude to test a feature, capture a bug, or run a security scan — all against your real app, with your real session, without any setup.

Quick Start Summary

# 60-second setup
git clone https://github.com/vitalysim/browser-bridge.git && cd browser-bridge
( cd server && npm install && npm run build ) && ( cd extension && npm install && npm run build )
( cd server && npm run install-service )
# Load extension in chrome://extensions, paste token, connect Claude Code

Then: claude "Navigate to my app and run a security audit"


Source: github.com

Sources cited in this article

  1. DevTools
Source: gentic.news · · author= · citation.json

AI-assisted reporting. Generated by gentic.news from 1 verified source, fact-checked against the Living Graph of 4,300+ entities. Edited by Ala SMITH.

Following this story?

Get a weekly digest with AI predictions, trends, and analysis — free.

AI Analysis

For Claude Code users, Browser Bridge changes how you approach any task that involves a web browser. Instead of manually copying data from authenticated apps into your Claude session, you can now delegate the entire interaction. The key workflow change: **add Browser Bridge as a standard MCP server in your CLAUDE.md** so every session has browser access by default. Start with the most common use cases: reading dashboards, capturing API traffic during debugging, and recording sessions for bug reports. The 63-tool surface means you can chain multiple browser operations — navigate to a page, extract data, interact with forms, capture network traffic — all in a single Claude Code command. This is especially powerful when combined with Claude's ability to write and test code against live APIs. Security consideration: because Browser Bridge gives your agent full browser access, only use it with trusted AI models. Claude Opus 5's demonstrated 0% prompt injection rate in browser agents makes it a safer choice than alternatives for this use case. Also, consider adding a `~/.browser-bridge/token` rotation strategy if you share your machine.
Compare side-by-side
Claude Code vs Browser Bridge MCP
Enjoyed this article?
Share:

AI Toolslive

Five one-click lenses on this article. Cached for 24h.

Pick a tool above to generate an instant lens on this article.

Related Articles

From the lab

The framework underneath this story

Every article on this site sits on top of one engine and one framework — both built by the lab.

More in Open Source

View all