Featured

Deploy OpenClaw in 60 seconds — 20% off logoDeploy OpenClaw in 60 seconds — 20% off

Launch OpenClaw on Hostinger in about 60 seconds and keep your agent live 24/7. Our referral link gives you 20% off, no coupon code needed.

Launch on Hostinger
Run your Hermes agent on Hostinger, fully managed logoRun your Hermes agent on Hostinger, fully managed

Launch Hermes on Hostinger in one click, fully managed, no VPS knowledge needed. Use code ZACAARON10 for 10% off.

Launch on Hostinger
Crawl and scrape any site into clean data, 10% off logoCrawl and scrape any site into clean data, 10% off

Firecrawl crawls and scrapes any site into clean markdown for your agent. Get 1,000 free credits, and new users get 10% off their first purchase.

Try Firecrawl free
6,000+ web scrapers for your AI agent, start free logo6,000+ web scrapers for your AI agent, start free

Apify gives your agent live web data: 6,000+ prebuilt scrapers and actors, MCP-ready. Sign up free with $5 in usage credits.

Try Apify free
One API to scrape, enrich, and extract the internet. logoOne API to scrape, enrich, and extract the internet.

Context.dev gives your agents a single API to scrape, enrich, and extract live web data — no proxies, no parsers, no maintenance.

Start building free
SetupClaw: done-for-you OpenClaw for founders & exec teams logoSetupClaw: done-for-you OpenClaw for founders & exec teams

White-glove OpenClaw for founders and exec teams (4–50+ employees): we install, harden, integrate your tools, and maintain it — secured from day one.

Get it set up for you
SEO data APIs for your agent, $1 free credit logoSEO data APIs for your agent, $1 free credit

DataForSEO gives your agent live access to SERP results, keyword data, backlinks, and on-page SEO data through one API. New accounts get a $1 credit, good for up to 20,000 keyword or backlink lookups.

Try DataForSEO free
Reach 48,000+ AI builders

A flat monthly placement in front of developers actively installing AI tools. No lock-in, cancel anytime.

Advertise here

Works with

Claude CodeClaude DesktopCursorVS CodeClineCodex CLIOpenClaw+ any MCP client

Install to Claude Code

This server doesn't publish a one-line install command. Follow the setup in the source repository.

Summary

Custom hosts mapping for Playwright MCP server, enabling browser automation to resolve hostnames to arbitrary IPs without modifying system /etc/hosts.

README.md

playwright-mcp-hosts

中文文档

Like SwitchHosts for Playwright MCP — custom hosts mapping for browser automation without modifying system /etc/hosts.

Installation

No manual installation required. Run on-demand via npx.

Configure MCP Client

Add the following to your MCP client config (e.g. ~/.cursor/mcp.json):

{
  "mcpServers": {
    "playwright": {
      "command": "npx",
      "args": ["playwright-mcp-hosts@latest"]
    }
  }
}

On first run, the config directory ~/.playwright-mcp/ and a default hosts.json will be created automatically.

Config Lookup Order

Config uses a project-first strategy (similar to .editorconfig):

  1. Project-level: $CWD/.playwright-mcp/hosts.json — applies to current project only
  2. Global: ~/.playwright-mcp/hosts.json — shared across all projects

If a project-level config exists, it takes full precedence and the global config is ignored.

Note: Project-level config relies on the MCP process working directory (process.cwd()) pointing to the project root. To use project-level config, create .cursor/mcp.json in your project (not in the global ~/.cursor/mcp.json), so that Cursor launches the MCP server with the project directory as the working directory.

Project-Level Config Example

  1. Create .cursor/mcp.json in the project root:
{
  "mcpServers": {
    "playwright": {
      "command": "npx",
      "args": ["playwright-mcp-hosts@latest"]
    }
  }
}
  1. Create .playwright-mcp/hosts.json in the project root (consider adding it to .gitignore).
  1. Restart the playwright MCP server in Cursor Settings → MCP.

Edit Hosts Config

  • Global: edit ~/.playwright-mcp/hosts.json
  • Project: edit <project>/.playwright-mcp/hosts.json

Restart the MCP server after editing.

Architecture

Browser requests www.example.com
  → CONNECT 127.0.0.1:18999 (local proxy)
  → Lookup hosts map: www.example.com → 10.0.0.1
  → TCP connect to 10.0.0.1:443
  → Browser completes TLS handshake with target
  → Page loads from the target environment

How It Works

Chromium 146+'s --host-resolver-rules is unreliable (ignored when sharing network service process with system browser). This tool uses a local CONNECT proxy with Playwright MCP's --proxy-server option instead:

  1. launch.mjs — Launcher script: loads hosts config, starts proxy, launches Playwright MCP
  2. proxy.mjs — Lightweight HTTP CONNECT proxy that redirects connections at the TCP level
  3. hosts.json — User config file (project-level or global), supporting remote / local / inline host sources

Dependencies

This package does not bundle @playwright/mcp as a hard dependency. It fetches the latest version on-demand via npx @playwright/mcp@latest at runtime.

Configuration

Full hosts.json example:

{
  "sources": [
    {
      "name": "staging",
      "type": "remote",
      "enabled": true,
      "url": "http://10.0.0.1/hosts",
      "timeout": 10000
    },
    {
      "name": "local hosts file",
      "type": "local",
      "enabled": false,
      "path": "~/.playwright-mcp/test-hosts.txt"
    },
    {
      "name": "inline mappings",
      "type": "inline",
      "enabled": false,
      "mappings": {
        "api.example.com": "10.0.0.100",
        "www.example.com": "10.0.0.101"
      }
    }
  ],

  "ignore_https_errors": true,
  "browser": "msedge",
  "extra_args": []
}

Source Fields

| Field | Description | |-------|-------------| | name | Label for log output | | type | remote (URL) / local (file) / inline (key-value mappings) | | enabled | true to enable / false to disable, like SwitchHosts toggles | | url | Remote URL returning a standard hosts file (remote type only) | | path | Local hosts file path, supports ~ (local type only) | | mappings | { "hostname": "ip" } object (inline type only) | | timeout | Fetch timeout in milliseconds, default 5000 (remote type only) |

Global Options

| Field | Description | Default | |-------|-------------|---------| | ignore_https_errors | Ignore HTTPS certificate errors (usually needed for staging) | false | | browser | Browser type: chrome / msedge / firefox / webkit | chrome | | proxy_port | Local proxy port for hosts mapping | 18999 | | extra_args | Additional arguments passed to Playwright MCP | [] |

Tip: Use msedge — When system Chrome is running, Playwright's Chrome instance may share its network service process, causing proxy settings to be ignored. Edge runs as a separate process and is not affected.

Usage

Switch Environments

Edit hosts.json, set the target environment's enabled to true and others to false, then restart the MCP server.

{
  "sources": [
    { "name": "staging", "enabled": true, ... },
    { "name": "pre-production", "enabled": false, ... }
  ]
}

Disable Hosts Mapping

Set all sources' enabled to false and restart. The browser will connect directly to production (no proxy).

Multiple Sources

Multiple sources can be enabled simultaneously. Later rules override earlier ones for the same hostname.

Remote Hosts File Format

Remote URLs should return content in standard hosts file format:

# comment
10.0.0.1  www.example.com
10.0.0.1  api.example.com
10.0.0.2  internal.example.com

See related servers & alternatives →

Related MCP servers

Browse all →

Related guides

Hand-picked reading to help you choose and use Browser & Scraping servers.