browser-gateway
Reliable, scalable browser infrastructure for AI agents and automation.
Route, pool, and failover across any browser provider. Built-in MCP server for AI agents.
What It Does
┌─────────────────────┐
│ browser-gateway │
│ │
│ routing / failover │
│ load balancing │
│ health monitoring │
│ request queuing │
└──────────┬───────────┘
│
┌──────────────────┼──────────────────┐
│ │ │
┌──────┴──────┐ ┌──────┴──────┐ ┌───────┴──────┐
│ Provider A │ │ Provider B │ │ Provider C │
│ Cloud CDP │ │ Playwright │ │ Local Chrome │
│ priority: 1 │ │ Docker :4000│ │ Docker :9222 │
└─────────────┘ └─────────────┘ └──────────────┘
One endpoint. Multiple providers. Automatic failover if one goes down.
Your app connects to ws://gateway:9500/v1/connect. The gateway picks the best available provider based on health, capacity, and your routing strategy. Providers can be cloud CDP services, Docker containers, or local Chrome instances.
Core Features
Routing & Reliability
- Automatic Failover - Provider down? Next one picks up instantly. Zero client changes.
- 5 Load Balancing Strategies - Priority chain, round-robin, least-connections, latency-optimized, weighted
- Per-Provider Concurrency Limits - Set
maxConcurrent per provider, gateway enforces it
- Request Queue - All providers busy? Connections wait instead of failing
- Cooldown System - Skip failing providers automatically, recover after TTL
- Health Checks - Periodic connectivity probes detect unhealthy providers
- Graceful Shutdown - Active sessions drain cleanly on SIGTERM/SIGINT
- Session Reconnect - If a client drops, reconnect with the same session ID and the gateway routes you back to the same provider. The browser there is still alive (as long as the provider keeps it running), so cookies, localStorage, and page state are intact. Works best with raw Chrome and providers that don't kill the browser on disconnect.
- Webhooks - Get notified when providers go down, recover, or queue overflows
REST API
- Screenshot -
POST /v1/screenshot — capture any page as PNG or JPEG
- Content Extraction -
POST /v1/content — get page content as markdown, text, HTML, or cleaned article
- Scrape -
POST /v1/scrape — extract data with CSS selectors or get full-page content
- Session Pool - Browser connections are reused across requests (like a database connection pool)
- Auto-Retry - Failed requests automatically retry with a fresh browser page
MCP Server for AI Agents
- 8 Browser Tools - navigate, snapshot, screenshot, viewport, interact, evaluate, close, status
- Zero Config - Auto-detects Chrome, launches on first tool use
- Concurrent Sessions - Multiple agents get separate browsers, no conflicts
- Raw CDP - Lightweight, no Playwright or Puppeteer dependency
- Works with Claude Code, Cursor, and any MCP-compatible client
Management
- Web Dashboard - Manage providers, view sessions, edit config from the browser
- Provider CRUD - Add, edit, delete, and test providers from the dashboard or API
- Config Editor - Edit gateway.yml with syntax highlighting and validation
- Auth - Token-based with secure HttpOnly cookie for the dashboard
- Protocol Agnostic - Works with Playwright, Puppeteer, any WebSocket protocol
Quick Start
As a WebSocket Proxy (for applications)
npm install -g browser-gateway
Create gateway.yml:
version: 1
providers:
primary:
url: wss://provider.example.com?token=${PROVIDER_TOKEN}
limits:
maxConcurrent: 5
priority: 1
fallback:
url: ws://my-playwright-server:4000
limits:
maxConcurrent: 10
priority: 2
browser-gateway serve
Connect from your app:
// For CDP providers
const browser = await chromium.connectOverCDP('ws://localhost:9500/v1/connect');
// For Playwright run-server providers
const browser = await chromium.connect('ws://localhost:9500/v1/connect');
Or use the REST API — no WebSocket management needed:
# Screenshot
curl -X POST http://localhost:9500/v1/screenshot \
-H "Content-Type: application/json" \
-d '{"url": "https://example.com"}' --output screenshot.png
# Extract content as markdown
curl -X POST http://localhost:9500/v1/content \
-H "Content-Type: application/json" \
-d '{"url": "https://example.com", "formats": ["markdown"]}'
Dashboard at http://localhost:9500/web.
As an MCP Server (for AI agents)
Add to your Claude Code or Cursor config:
{
"mcpServers": {
"browser-gateway": {
"command": "npx",
"args": ["browser-gateway", "mcp"]
}
}
}
No config files needed. The agent can now browse websites, take screenshots, fill forms, and extract data.
See the MCP documentation for all options.
Dashboard
Built-in web dashboard at http://localhost:9500/web. Served from the same port as the gateway.
| Page | What You Can Do |
|---|
| Overview | Gateway health at a glance: active sessions, provider status, connection endpoint |
| Providers | Add, edit, delete, and test browser providers. Changes write to gateway.yml |
| Sessions | Live table of every active connection: provider, duration, message count |
| Config | Edit gateway.yml in the browser with validation and automatic backups |
If BG_TOKEN is set, the dashboard requires authentication via a secure HttpOnly cookie.
See Dashboard Guide for details.
Authentication
Set BG_TOKEN to require a token (or put it in a .env file):
BG_TOKEN=my-secret-token browser-gateway serve
- WebSocket clients pass the token as
?token= query param
- API clients use
Authorization: Bearer <token> header
- Dashboard shows a login form, sets a secure HttpOnly cookie
- Health endpoint (
/health) is always public
CLI
# Proxy server
browser-gateway serve # Start the gateway + dashboard
browser-gateway serve --port 8080 # Custom port
browser-gateway serve --config path.yml # Custom config
# MCP server for AI agents
browser-gateway mcp # Auto-detect Chrome, zero config
browser-gateway mcp --headless # Headless mode (for CI/Docker)
browser-gateway mcp --cdp-endpoint ws:// # Connect to existing browser
browser-gateway mcp --config gateway.yml # Multi-provider with failover
# Utilities
browser-gateway check # Test provider connectivity
browser-gateway version # Print version
browser-gateway help # Show help
API
| Endpoint | Method | Description |
|---|
/v1/connect | WebSocket | Connect to a browser (the core feature) |
/v1/screenshot | POST | Take a screenshot of any URL (docs) |
/v1/content | POST | Extract page content as markdown, text, or HTML (docs) |
/v1/scrape | POST | Extract data via CSS selectors or full-page formats (docs) |
/v1/status | GET | Gateway health + provider status + pool status |
/v1/sessions | GET | Active sessions |
/v1/providers | GET/POST | List or add providers |
/v1/providers/:id | PUT/DELETE | Update or remove a provider |
/v1/providers/:id/test | POST | Test provider connectivity |
/v1/config | GET/PUT | Read or save config |
/v1/config/validate | POST | Validate YAML without saving |
/mcp | POST | MCP Streamable HTTP endpoint |
/json/version | GET | CDP discovery (for browser-use, Playwright, Stagehand) |
/health | GET | Health check |
Docker
docker run -d \
-p 9500:9500 \
-v ./gateway.yml:/app/gateway.yml:ro \
-e PROVIDER_TOKEN=xxx \
ghcr.io/browser-gateway/server:latest
How It Works
- Client connects to
ws://gateway:9500/v1/connect
- Gateway selects a provider using your routing strategy
- Gateway opens a raw TCP connection to the provider
- HTTP upgrade forwarded, provider responds with
101 Switching Protocols
- Bidirectional TCP pipe:
client <-> gateway <-> provider
- All WebSocket messages forwarded transparently (never parsed or modified)
- On disconnect: session cleaned up, slot released, metrics updated
- If all providers full: connection waits in a queue until a slot opens
Works With
browser-gateway is compatible with existing browser tools. Just pass the gateway URL — it auto-resolves via /json/version.
AI Agent Frameworks:
# browser-use (Python) — HTTP URL auto-resolves
BrowserSession(cdp_url="http://localhost:9500")
// Stagehand (TypeScript)
new Stagehand({ env: "LOCAL", localBrowserLaunchOptions: { cdpUrl: "http://localhost:9500" } })
Playwright MCP (all 70 Playwright tools through gateway routing):
{
"mcpServers": {
"playwright": {
"command": "npx",
"args": ["@playwright/mcp@latest", "--cdp-endpoint", "http://localhost:9500"]
}
}
}
Puppeteer / Playwright:
// Playwright — HTTP or WebSocket
const browser = await chromium.connectOverCDP("http://localhost:9500");
// Puppeteer — WebSocket
const browser = await puppeteer.connect({ browserWSEndpoint: "ws://localhost:9500/v1/connect" });
Documentation
Contributing
Contributions welcome. See CONTRIBUTING.md for guidelines.
License
MIT - see LICENSE.
Links
