Provides tools for controlling Android devices via ADB and scrcpy, including device management, input injection, screen capture/recording, app management, clipboard operations, file transfer, and mirroring.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@scrcpy MCP Server take a screenshot of my connected device"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Local MCP server that exposes Android device control via adb and scrcpy as 22 tools for MCP clients like Claude Code and Codex CLI.
One command — no prerequisites needed (the installer handles everything):
curl -fsSL https://raw.githubusercontent.com/charettep/scrcpy-mcp/main/install.sh | bashOr from a local clone:
git clone https://github.com/charettep/scrcpy-mcp.git && cd scrcpy-mcp
./install.shThe installer will:
Detect your OS and package manager (apt, dnf, pacman, brew)
Download the repo to ~/mcp/scrcpy-mcp (if running via curl)
Check for adb, scrcpy, and uv — install any that are missing
Find Python 3.10+ and create an isolated venv with dependencies
Detect installed MCP clients (Claude Code, Codex CLI)
Ask where to register the server:
Global — injects config directly into ~/.claude.json and/or ~/.codex/config.toml
Project — generates a local .mcp.json file
Re-running the installer is safe — it updates the existing install and overwrites old config entries.
Related MCP server: DeepADB
Python 3.10+
adb (from Android SDK platform-tools)
scrcpy (for mirroring/recording sessions)
An Android device with USB debugging enabled
The installer handles all of the above automatically.
If you prefer not to use the installer:
uv venv .venv && uv pip install -r requirements.txtThen add to your project .mcp.json:
{
"mcpServers": {
"scrcpy": {
"command": "/path/to/.venv/bin/python",
"args": ["/absolute/path/to/scrcpy_mcp.py"],
"env": {
"SCRCPY_MCP_ADB_PATH": "/usr/bin/adb",
"SCRCPY_MCP_SCRCPY_PATH": "/usr/bin/scrcpy"
}
}
}
}The env block is optional — without it, adb and scrcpy are resolved from PATH.
Tool | Description |
| List connected devices with status |
| Get model, Android version, screen resolution, battery, IP |
| Connect/disconnect device via wireless ADB |
Tool | Description |
| Tap at x,y coordinates |
| Swipe from point A to B with duration |
| Send key event (BACK, HOME, POWER, etc.) |
| Type text string on device |
| Long press at coordinates |
Tool | Description |
| Capture screenshot and save locally |
| Turn screen on/off/toggle |
| Get/set device rotation |
| Record screen to mp4 file |
Tool | Description |
| List installed packages |
| Launch app by package name |
| Force stop an app |
| Install APK file |
Tool | Description |
| Read device clipboard |
| Set device clipboard text |
Tool | Description |
| Push local file to device |
| Pull file from device |
Tool | Description |
| Start scrcpy mirroring with full options |
| Stop a running scrcpy session |
All tools accept an optional serial parameter. If omitted, adb uses the only connected device (or fails if multiple are connected).
tap(x=500, y=1000, serial="ABCD1234")Variable | Purpose | Default |
| Absolute path to |
|
| Absolute path to |
|
These are set automatically by install.sh via the MCP config env block.
python3 scrcpy_mcp.py # starts on stdio
python3 -c "from scrcpy_mcp import mcp; print(len(mcp._tool_manager._tools))" # should print 22This server cannot be deployed
Control Android TV from any AI. 38 MCP tools: playback, recap, recommend, smart-home, schedules.
Melaya is a remote MCP server. It gives an assistant hands on your own Android phone and browser: it reads the screen through the accessibility tree, then taps, types and navigates inside the apps and sites you allow-list, with no per-app API. It also builds, schedules and runs agent pipelines across 6k+ connected tools. OAuth 2.1, nothing to install.
Control real Android and iOS devices with LLM agents — tap, swipe, type, automate flows.
Drive real devices from your AI Coding tool. Embed a client SDK (Unity, Godot, Flutter, iOS/macOS, Android, React Native, Web) in your app, then capture screenshots, traverse the UI tree, inject taps and key events, and run automated test tasks on the physical device over a secure relay.
