Skip to main content
Glama

macuse

by macuse-app

Macuse


Macuse is a dual‑mode desktop application that bridges AI assistants with native macOS functionality through the Model Context Protocol (MCP). It enables AI clients to control Mac applications, access system information, and automate workflows while maintaining security and user privacy.

Key Features

Macuse provides practical, local capabilities through toolboxes, including:

  • System: launch and manage apps, general system operations
  • Inspector: explore app UI structure to understand elements
  • Interaction: click/type/interact with UI elements
  • Calendar & Reminders: read and manage events and reminders (with consent)
  • Contacts, Notes, Messages: access common personal apps (with consent)
  • Location & Maps: read current location and use mapping features (with consent)

Everything runs locally. Access is protected by a token you control, and sensitive actions require the appropriate macOS permissions.

Installation

  1. Download the latest installer from the Website: https://macuse.app/download/
  2. Unzip the downloaded file.
  3. Drag Macuse.app into your Applications folder.

Configure Macuse

Open Macuse to review and adjust your connection and capabilities:

  • Access Token: copy your token to connect clients, you can regenerate it anytime.
  • Transport Mode:
    • Stdio: direct connection with no open ports, simplest for most clients.
    • SSE: runs a local server on your bind address/port and streams events.
    • Streamable HTTP: runs a local HTTP endpoint suitable for clients that prefer HTTP.
  • Network (for SSE/HTTP):
    • Bind Address: default 127.0.0.1 (local only).
    • Port: default 35729. Use the built‑in port check to avoid conflicts.
  • Toolboxes: enable/disable capabilities such as Inspector, Interaction, Calendar, Contacts, Messages, Notes, and Location.
  • Permissions: macOS will prompt for Accessibility, Contacts, Calendars, Reminders, Location, etc., when needed.

Connect Your MCP Client

  1. In Macuse:
  • Choose your Transport Mode (Stdio recommended). If using SSE/HTTP, confirm bind address and port.
  1. In your client:
  • If the client supports .mcpb plugins
    • Copy your Access Token.
    • Download and install macuse.mcpb, then provide your Access Token (and binary path if prompted; default is /Applications/Macuse.app/Contents/MacOS/macuse).
  • If the client does not support .mcpb
    • Use Macuse’s one‑click setup buttons to configure supported clients (e.g., Claude, Cursor, Raycast, AnythingLLM), or
    • Copy the generated configuration from Macuse and paste it into your client’s settings for manual setup.

Manual Configuration (JSON Examples)

Generic connection config (copy/paste friendly):

Stdio

{ "mode": "stdio", "config": { "type": "stdio", "command": "/Applications/Macuse.app/Contents/MacOS/macuse", "args": ["--stdio"], "env": { "ACCESS_TOKEN": "<YOUR_TOKEN>" } } }

SSE

{ "mode": "sse", "config": { "type": "sse", "url": "http://<BIND_ADDRESS>:<PORT>/sse", "headers": { "Authorization": "Bearer <YOUR_TOKEN>" } } }

Streamable HTTP

{ "mode": "streamableHttp", "config": { "type": "streamable-http", "url": "http://<BIND_ADDRESS>:<PORT>/mcp", "headers": { "Authorization": "Bearer <YOUR_TOKEN>" } } }

Privacy & Security

  • The Access Token authenticates your client to Macuse locally; it is not sent to remote servers.
  • macOS permissions (Contacts, Calendar, Accessibility, etc.) must be explicitly granted by you when needed.
  • Website: https://macuse.app/
  • Download: https://macuse.app/download/
  • Model Context Protocol: https://modelcontextprotocol.io/
-
security - not tested
F
license - not found
-
quality - not tested

local-only server

The server can only run on the client's local machine because it depends on local resources.

Connect AI with any macOS app. Deep integration with native apps like Calendar, Mail, Notes, plus UI control for all applications. Works with Claude, Cursor, Raycast, and any MCP-compatible AI.

  1. Key Features
    1. Installation
      1. Configure Macuse
        1. Connect Your MCP Client
          1. Manual Configuration (JSON Examples)
        2. Privacy & Security
          1. Helpful Links

            Related MCP Servers

            • A
              security
              A
              license
              A
              quality
              A collection of tools that enables Claude AI and Cursor to access native macOS applications such as Messages, Notes, Contacts, Emails, Reminders, Calendar, and Maps through the Model Context Protocol.
              Last updated -
              7
              262
              11
              MIT License
              • Apple
            • A
              security
              F
              license
              A
              quality
              A collection of tools that enable AI assistants to interact with Apple applications and services including Contacts, Notes, Messages, Mail, Reminders, Calendar, Maps, and Web Search through the Model Context Protocol.
              Last updated -
              8
              1
              • Apple
            • -
              security
              A
              license
              -
              quality
              An MCP server that enables AI assistants like Claude to access and manipulate Apple Notes on macOS, allowing for retrieving, creating, and managing notes through natural language interactions.
              Last updated -
              75
              MIT License
              • Apple
            • -
              security
              A
              license
              -
              quality
              A collection of Apple-native tools for the MCP protocol that enables AI assistants to interact with Apple applications including Messages, Notes, Contacts, Mail, Reminders, Calendar, and Maps.
              Last updated -
              262
              MIT License
              • Apple

            View all related MCP servers

            MCP directory API

            We provide all the information about MCP servers via our MCP API.

            curl -X GET 'https://glama.ai/api/mcp/v1/servers/macuse-app/macuse'

            If you have feedback or need assistance with the MCP directory API, please join our Discord server