PhonePi MCP: Turn your phone into a toolbox for AI assistants using Model Context Protocol

Name: PhonePi MCP: Turn your phone into a toolbox for AI assistants using Model Context Protocol
Author: priyankark

@priyankark

Repository Homepage

About PhonePi MCP: Turn your phone into a toolbox for AI assistants using Model Context Protocol

PhonePi MCP enables seamless integration between desktop AI tools and your smartphone, providing 23+ direct actions including SMS messaging, phone calls, contact management, snippet creation and search, clipboard sharing, notifications, battery status checks, and remote device co

#remote-control #ai #mcp #cursor #claude #remote-access-tool #phone-link #mcp-server #modelcontextprotocol #mcp-client #phonepi-mcp #phone-mcp #mcp-phone-control #phone-pc-link #phone-p2p #ai-smartphone #mobile-mcp

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "phonepi-mcp": {
      "command": "npx",
      "args": [
        "phonepi-mcp",
        "start"
      ]
    }
  }
}

Tools

Get the current battery level of the phone

Add a new snippet to the phone

Get all snippets from the phone

Get a specific snippet by ID

Get snippets by type

Search snippets by query

Update an existing snippet

Delete a snippet from the phone

Send an SMS message

Make a phone call

Make the phone beep to help locate it

Set a timer for a specific duration

Copy text to the phone clipboard

Get all contacts from the phone

Get a specific contact by ID

Add a new contact to the phone

Update an existing contact

Delete a contact from the phone

Share a snippet via messaging apps

Send a message to the phone that requires user attention or response

Get the response to a previously sent message

Send a notification to the phone

Overview

What is PhonePi MCP?

PhonePi MCP is an open-source MCP server and companion app that lets you manage and control your Android phone remotely using natural language commands via MCP-supported AI assistants like Cursor and Claude Desktop. It provides a comprehensive set of programmatic tools to interact with your mobile device.

How to use PhonePi MCP?

You host the MCP server locally (source code is fully open source) and install the companion app on your phone. The app and server communicate over your local network; no third-party remote servers are involved. For detailed instructions, visit the project’s official website.

Key features of PhonePi MCP

Manage contacts and messages
Handle snippets and notes
Control phone settings and notifications
Send SMS and make calls
Share content across apps
Monitor battery status
Set timers and reminders
Find your phone with audio alerts

Use cases of PhonePi MCP

Have an AI assistant send a text message or make a call on your behalf.
Let an AI tool check your phone’s battery level and notify you when low.
Remotely set timers or reminders using natural language commands.
Quickly share a URL or note from your desktop to your phone.
Trigger an audio alarm on your phone to find it in the house.

FAQ from PhonePi MCP

How does PhonePi MCP ensure security?

You control the server (100% open source), decide which app permissions to grant, and the connection happens over your local network with no third-party remote servers. It is recommended to use a secure, private network (e.g., Tailscale) and avoid public Wi-Fi.

Which AI assistants are supported?

PhonePi MCP works with any MCP‑supported AI client, with Cursor and Claude Desktop explicitly mentioned.

Do I need internet access to use PhonePi MCP?

The server and app communicate over your local network, so no external internet is required for the connection itself. However, your phone may need internet for features like sending SMS or making calls.

Can I use PhonePi MCP on a public network?

It is not recommended. The README advises against using public networks like airports and restaurants; a private VPN (such as Tailscale) is strongly recommended.

What data is sent to third parties?

No data is sent to third parties. The server runs locally on your machine, and the app connects directly to it over your local network. You control all permissions.