home / mcp / everything search mcp server

Everything Search MCP Server

Provides an MCP bridge to the Everything Search Engine for fast, natural-language file searches across the system.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "ananyaakamat-everything-search-mcp-server": {
      "command": "everything-search-mcp-server",
      "args": []
    }
  }
}

You can use the Everything Search MCP Server to connect MCP-compatible clients to the Everything Search Engine, enabling fast, natural-language file searches across your system. This server exposes a simple, local command interface so you can query files, filter results, and integrate search into tools like Claude Desktop or VS Code.

How to use

Start the MCP server locally and connect your MCP client to it. You will run the MCP server and then configure your client to point at the server so you can perform fast file searches, apply advanced filters, and see results rendered in your client.

How to install

Prerequisites you need before installation include Node.js and the Everything Search Engine runtime running on Windows.

1. Install Everything Search Engine on Windows and ensure its HTTP server is enabled on port 8011.

2. Install the MCP server globally or from source.

3. Configure your MCP client to use the server, then restart the client to load the new configuration.

# Option A: Global installation (recommended)
npm install -g everything-search-mcp-server

# Option B: From source
git clone https://github.com/ananyaakamat/everything-search-mcp-server.git
cd everything-search-mcp-server
npm install
npm run build
npm install -g .

Configuration and usage notes

Your MCP client needs to know how to start the MCP server. The recommended approach uses a simple stdio (local) setup where the client launches the MCP server binary directly.

{
  "type": "stdio",
  "name": "everything_search",
  "command": "everything-search-mcp-server",
  "args": []
}

Available tools

search

Advanced file search that supports scope, case sensitivity, whole word matching, regex, path focus, and sorting.

natural_search

Natural language search to describe queries like large video files or recent documents.