# lanchuske/local-mcp [Health: Active]

**Category:** 💬 Communication  
**Repository:** https://github.com/lanchuske/local-mcp-releases  
**GitHub Stars:** 79  
**npm Downloads (last month):** 15794  
**Views:** 12  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/lanchuske-local-mcp

## Description
Connect Claude, Cursor, Windsurf and other AI agents to macOS native apps: Mail, Calendar, Contacts, Reminders, Notes, iMessage, Finder, Safari, OmniFocus, Microsoft Teams, Outlook, OneDrive, and Office documents. 82 tools. Runs entirely on your Mac — no cloud, no tokens, no API keys.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "local-mcp": {
    "command": "npx",
    "args": ["-y","local-mcp@latest"]
  }
}
```

## Documentation

## What lanchuske/local-mcp does

lanchuske/local-mcp gives compatible AI clients MCP access to information and actions across desktop applications, connected accounts, and local or synchronized files. Its integrations cover areas such as email, calendars, contacts, messages, reminders, notes, tasks, file storage, browser automation, and Office documents. Examples named by the project include Apple Mail, Outlook, Microsoft Teams, Slack, WhatsApp, Signal, Google Drive, OneDrive, Microsoft 365, OmniFocus, Notion, and ServiceNow.

The main use case is cross-application context. An assistant can search related messages, email, calendar entries, and files, then use available tools to draft content, organize information, or make supported updates after authorization. The exact tools and integrations depend on the operating system, connected accounts, permissions, and client.

## How it works

The installed LMCP application runs the MCP tools on the user's computer. A compatible assistant sends requests through MCP; LMCP selects the available application or account integration and returns context or performs an authorized action. Some integrations read native data, local caches, or synchronized files. Others require a network connection and sign-in.

Desktop and command-line clients generally launch the server through the `local-mcp` npm wrapper or an installed integration. Web assistants use `https://www.local-mcp.com/mcp` as a custom connector with OAuth. That route requires the local application and Cloud Data Forwarding; forwarding is disabled by default, and without it the connector exposes only `setup_install`.

## Setup and configuration

Install LMCP for macOS or Windows first, then select the assistant and integrations to use. The supported platforms are macOS 13 or later and 64-bit Windows 10 or later. Required operating-system permissions and service sign-ins must be granted during setup.

For a Node.js-based setup, run:

```bash
npx -y local-mcp@latest setup
```

The setup command configures compatible clients it detects. A client can also be configured directly with the `npx -y local-mcp@latest` command and its corresponding MCP settings format. Cursor offers both the setup command and a repository plugin. A Claude Desktop `.mcpb` extension, signed installers, and platform-specific installation commands are also available.

Restart the AI client after installation because MCP tools load at startup. For web assistants, complete connector authorization and verify that tools are available; an endpoint by itself does not confirm a working connection.

## Tools and capabilities

Available capabilities include:

- Search and work with email, messages, contacts, calendars, reminders, notes, and tasks.
- Access files and connected storage such as OneDrive and Google Drive.
- Work with Office documents and Microsoft 365 services.
- Use browser automation and supported productivity applications.
- Combine information from multiple apps to support planning, drafting, and follow-up actions.
- Apply app-specific permissions and account access rather than granting universal access automatically.

The project directs users to its live tool reference and setup guides for current tool names, requirements, and platform coverage.

## Limitations and notes

Integration availability differs between macOS and Windows. Apple-native tools require macOS, while connected services may need their own authentication and permissions. Some integrations operate on cached or synchronized data rather than a complete cloud account.

Assistant behavior and tool permissions still determine what can be read or changed. Demonstrated workflows are examples of possible tool combinations, not guarantees of autonomous results. The core binary is proprietary, while the repository documentation and scripts have a separate license.

_Full upstream README: https://allmcps.com/mcp/lanchuske-local-mcp/readme_