mcp
Browser MCP is a TypeScript-based server that lets AI applications (Claude, Cursor, VS Code) control your browser through automation. It uses your existing browser profile and runs locally, keeping your activity private and avoiding bot detection.
Key facts
Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.
| Field | Value |
|---|---|
| Repository | BrowserMCP/mcp |
| Owner | BrowserMCP |
| Primary language | TypeScript |
| License | Apache-2.0 — OSI-approved |
| Stars | 6.8k |
| Forks | 527 |
| Open issues | 141 |
| Latest release | Unknown |
| Last updated | 2025-04-24 |
| Source | https://github.com/BrowserMCP/mcp |
What mcp is
An MCP (Model Context Provider) server + Chrome extension enabling AI-driven browser automation via existing user sessions. Runs locally to eliminate network latency and preserve browser fingerprinting; adapted from Microsoft's Playwright MCP but targets user browsers rather than headless instances.
Get the mcp source
Clone the repository and explore it locally.
git clone https://github.com/BrowserMCP/mcp.gitcd mcp# follow the project's README for install & configurationNeed it deployed, integrated, or customized instead? DEV.co ships production installs.
Best use cases
Implementation considerations
- Project explicitly states it cannot yet be built standalone due to monorepo dependencies on internal utils and types—building/forking requires access to the full monorepo or substantial reverse-engineering.
- Requires Chrome extension installation and local MCP server setup; integration is IDE/AI-tool specific (Cursor, Claude, VS Code) and not universal.
- Session and profile management are tied to your local browser state; no built-in session isolation or multi-profile switching—suitable only for single-user or tightly controlled scenarios.
- No documented error handling, retry logic, or failure modes for long-running automation or network interruptions.
- Only ~2 months old with 141 open issues; breaking changes, API instability, and missing core features are likely.
When to avoid it — and what to weigh
- Centralized, Multi-User Automation at Scale — If you need to automate browser tasks across many users or machines, or require a remote SaaS solution, Browser MCP's local-only architecture and single-machine focus make it unsuitable.
- Headless or Unattended Server Deployment — Browser MCP requires a real Chrome browser instance with a user profile; it cannot run headless or in traditional server environments without display capability.
- Production-Grade, High-Availability Automation — No versioned releases or SLA. Project is <2 months old (created 2025-03-28); production dependencies on an early-stage, rapidly evolving codebase carry risk.
- Cross-Browser or Non-Chromium Scenarios — Currently Chrome-only. If you need Safari, Firefox, or edge browser support, this is not a fit.
License & commercial use
Licensed under Apache License 2.0 (Apache-2.0), a permissive OSI-approved license that permits commercial use, modification, and distribution under stated conditions.
Apache-2.0 permits commercial use. However, project documentation does not state warranty, SLA, or support terms. For production commercial deployments, ensure your use case and risk tolerance align with an early-stage, unsupported project that cannot be built standalone.
DEV.co evaluation signals
Editorial assessment — not user reviews. Directional, with an explicit confidence level.
| Signal | Assessment |
|---|---|
| Maintenance | Active |
| Documentation | Adequate |
| License clarity | Clear |
| Deployment complexity | High |
| DEV.co fit | Possible |
| Assessment confidence | Medium |
Project runs browser automation locally, preserving your session and avoiding data exfiltration to remote servers. However, (1) extension communicates with local MCP server over IPC—review IPC security if untrusted code is present; (2) MCP protocol and extension permissions should be audited for injection vectors; (3) no public security policy or reporting mechanism documented; (4) early stage implies security review is incomplete.
Alternatives to consider
Playwright MCP (Microsoft)
Official, well-maintained Microsoft MCP server for browser automation. Does not use user profile (creates fresh instances), so loses logged-in state but is more portable and production-ready.
Puppeteer / Playwright (programmatic)
Mature Node.js libraries for headless browser control. Require code integration rather than IDE plugin, but are widely battle-tested, versioned, and support CI/CD pipelines.
Selenium / WebDriver
Industry-standard, cross-browser automation with deep enterprise support. Heavier than MCP but offer robust error handling, multi-language bindings, and proven production deployments.
Build on mcp with DEV.co software developers
Browser MCP offers local, privacy-preserving browser automation for development and prototyping. However, production use requires careful evaluation due to its early maturity (4 weeks old), monorepo dependency, and lack of releases. Consult our full technical assessment and consider alternatives for mission-critical workflows.
Talk to DEV.coRelated open-source tools
Surfaced by semantic similarity across the DEV.co open-source index.
Related on DEV.co
Explore the category and the services that help you build with it.
mcp FAQ
Can I build Browser MCP from source?
Is Browser MCP suitable for production web automation?
Does Browser MCP work without a Chrome extension?
What AI tools are compatible with Browser MCP?
Custom software development services
DEV.co helps companies turn open-source tools like mcp into production software. Our software development services cover the full lifecycle — architecture, web development, integration, and maintenance — delivered by software developers and web developers who ship. Engage our software development agency to implement or customize it for your mcp servers stack.
Explore Browser MCP for Your AI Automation Needs
Browser MCP offers local, privacy-preserving browser automation for development and prototyping. However, production use requires careful evaluation due to its early maturity (4 weeks old), monorepo dependency, and lack of releases. Consult our full technical assessment and consider alternatives for mission-critical workflows.