BRYG
An MCP server that lets AI assistants drive real Chromium browsers — navigate, click, type, read pages, run OCR, and record network traffic. 43 tools, credentials stay local, zero telemetry.
README
<div align="center">
BRYG
Your AI assistant's hands on the browser
BRYG is an MCP server that lets AI assistants drive real Chromium-based browsers — navigate, click, type, read pages, run OCR, and record network traffic. Like a human operator, on your machine.
🌐 Website · 📖 Docs · 🔒 Security · 🔑 Privacy
$10/month · 14-day free trial · macOS & Windows · Works with Google Chrome
</div>
What is BRYG?
BRYG implements the Model Context Protocol (MCP) so any compatible client — such as Claude Code — can control a Chromium browser on your machine. It gives your AI assistant 43 tools covering perception, action, and observation, designed from daily QA work rather than demos.
Note: This repository hosts BRYG's public documentation and issue tracker. The product itself is distributed as a signed binary via the installer below.
Get started
BRYG is a licensed product — you need a key to activate it. It takes about a minute:
- Get a license key — start your 14-day free trial at bryg.pages.dev. Your license key is emailed to you (billing, cancellation, and refunds are handled in the customer portal linked in that email).
- Install — run the one-line installer for your OS (below).
- Activate — the guided setup asks for your key during install, or run
bryg activate <key>anytime.
Install
macOS
curl -fsSL https://bryg.pages.dev/install.sh | sh
Windows (PowerShell)
irm https://bryg.pages.dev/install.ps1 | iex
The installer downloads the right binary for your OS/architecture to ~/.bryg/bin/ (%USERPROFILE%\.bryg\bin\ on Windows), verifies its SHA-256 checksum, then runs a guided setup that registers BRYG with your MCP client(s), optionally installs OCR support, and activates your license key.
Why BRYG
- Sees pages like structure — accessibility-tree snapshots with stable element refs (
[ref=N]); compact and reliable for AI to act on, with coordinate fallback. - OCR & QR built in — reads text from canvas, images, and games (multi-strategy OCR, any tesseract language) and decodes QR codes on the page.
- Network capture — record everything, keyword-filtered, or failures-only. Sensitive headers and password/token-like fields are masked by default.
- Credentials stay local — secrets never enter AI conversations.
- Headful by design — a dedicated browser instance you can watch and take over at any time.
- Zero telemetry — runs entirely on your machine; the only network calls are license validation and an anonymous update check.
Tools (43)
Perception — browser_snapshot · browser_read · browser_screenshot · browser_ocr · browser_qr
Action — navigate · click · type · fill · scroll · press keys · drag · upload · select, and more
Observation — network_start · network_stop · network_status
See the full tool reference.
CLI
| Command | Description |
|---|---|
bryg |
Start the MCP server (stdio) — this is what your MCP client runs |
bryg setup [--yes] [--no-ocr] [--no-path] |
Guided setup (re-runnable) |
bryg activate <key> |
Activate your license on this machine |
bryg deactivate |
Release this machine's license seat |
bryg update [--check] |
Update BRYG (verified by checksum) |
bryg browser launch [url] | stop |
Manage the dedicated automation browser |
bryg uninstall [--yes] |
Deactivate seat, unregister, and remove ~/.bryg |
Documentation
Full documentation — installation, CLI, configuration, all 43 tools, licensing FAQ, and changelog — lives at bryg.pages.dev/docs.
Support
- 🐛 Found a bug or have a feature request? Open an issue.
- 💬 Questions? See the FAQ or reach out via the site.
License
BRYG is commercial software provided under its End User License Agreement. This repository's documentation content is © BRYG. The BRYG binary is not open source.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.