Server facts
Google Search Tool MCP Server
Project
- Author
- web-agent-master
- License
- Not specified
- Category
- Search & Web Data
Tech
- Runtime / language
- nodeTypeScript
- Transports
- stdio
- Remote available
- No
- Authentication
- Not specified
Activity
- GitHub stars
- 623
- Last commit
- Data as of
Tags
- Google Search
- Playwright
- Typescript
- Web Search
Overview
Google Search Tool gives AI assistants and command-line users a browser-driven way to query Google. Written in TypeScript for Node.js, it uses Playwright rather than a paid search-results API. Parsed responses contain result titles, destination links, and snippets in JSON. The same project includes a separate MCP entry point so an assistant such as Claude can request current search information. Users can adjust the result limit and timeout, or show the browser when investigating a problem. Browser fingerprint variation, stored session state, and switching between hidden and visible browsing are documented techniques for handling verification challenges.
For debugging search-page changes, the tool can retrieve HTML with scripts and styles removed instead of returning parsed results. Saving that HTML also creates a full-page screenshot, and an output-path option controls where the HTML goes. Setup requires Node.js 16 or later and Playwright browser binaries; source installation additionally involves dependency installation, compilation, and package linking. The README includes Windows-specific launch alternatives and troubleshooting guidance. It also warns against frequent requests, notes that some locations may need a proxy, and asks users to follow Google's policies. Saved browser state contains cookies and storage information, so it needs careful handling rather than being treated as a harmless log file.
Install & configure
Requirements
- Node.js 16 or later
- Playwright browser binaries
- An MCP-compatible client
npx google-search-mcpThe README uses google-search-mcp for MCP, distinct from the search CLI. Its source workflow installs dependencies, builds the project, and links the package globally. Browser binaries may download on first use. No required environment variables are stated.
Examples are translated from the published configuration. Supply your own credentials and verify requirements in the upstream README. Remote connectors are shown only where a supported configuration format is known.
Config file · macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Use Settings → Developer → Edit Config when available. Do not assume a Linux configuration path; check the client documentation for your platform. Remote servers use the app’s connector flow when supported, not the local stdio file shown here. Merge this example with existing settings; do not overwrite unrelated servers.
{
"mcpServers": {
"google-search": {
"command": "npx",
"args": [
"google-search-mcp"
],
"env": {}
}
}
}Config file · Project: .cursor/mcp.json
User: ~/.cursor/mcp.json
The project path is relative to the workspace root. The tilde denotes your user home directory. Use Cursor’s MCP settings to confirm which configuration and scope your installed version is reading. Merge this example with existing settings; do not overwrite unrelated servers.
{
"mcpServers": {
"google-search": {
"command": "npx",
"args": [
"google-search-mcp"
],
"env": {}
}
}
}Config file · Workspace: .vscode/mcp.json
For user scope, use the “MCP: Open User Configuration” command if available. Do not guess a user configuration path across operating systems, profiles, containers, or remote workspaces; check the client documentation. Merge this example with existing settings; do not overwrite unrelated servers.
{
"servers": {
"google-search": {
"type": "stdio",
"command": "npx",
"args": [
"google-search-mcp"
],
"env": {}
}
}
}Config file · macOS/Linux: ~/.codeium/windsurf/mcp_config.json
Windows: %USERPROFILE%\.codeium\windsurf\mcp_config.json
Open the MCP configuration through Windsurf’s settings when possible to confirm the active file. Paths can differ in managed or future releases; check the client documentation if this location is not present. Merge this example with existing settings; do not overwrite unrelated servers.
{
"mcpServers": {
"google-search": {
"command": "npx",
"args": [
"google-search-mcp"
],
"env": {}
}
}
}Config file · Project scope: .mcp.json at the repository root
The project file uses mcpServers. For user scope, use claude mcp add --scope user rather than assuming .mcp.json in your home directory is the user configuration. Check the client documentation for managed settings and storage details. Merge this example with existing settings; do not overwrite unrelated servers.
{
"mcpServers": {
"google-search": {
"type": "stdio",
"command": "npx",
"args": [
"google-search-mcp"
],
"env": {}
}
}
}Examples are translated from the published configuration. Supply your own credentials and verify requirements in the upstream README. Remote connectors are shown only where a supported configuration format is known.
Tools
No individual tool names were confirmed in the source material. Check the repository for its current tool list.
Names, permissions, and defaults can change. Check the current upstream documentation before granting access.
Use cases
- Give an MCP assistant current Google search results.
- Collect titles, links, and snippets from command-line queries.
- Capture search-page HTML and screenshots to investigate parsing changes.
- Inspect verification problems in a visible browser.
Strengths
- Searches do not require a paid SERP API key.
- Provides both CLI and MCP entry points.
- Includes HTML capture, screenshots, and browser-state reuse.
Limitations
- Google verification and request blocking remain operational concerns.
- Requires Playwright browser installation.
- Saved browser state contains sensitive session information.
Frequently asked questions
Independent listing, not an affiliation or security review. Grant the least access needed and verify the upstream instructions before running code.
