Drengr

Name: Drengr
Author: drengr

@drengrMCP server · External

Eyes and hands for AI agents on Android and iOS devices.

Computer controlv0.10.8Shell

Freeno charge per call

How to install

Runs on your machine from its package; the steps are below.

What it does

Reads
Writes
High impact

At most: moves money, runs code or commands, deletes data, or sends messages for you.

What it costs

Free

Odel charges nothing per call. The publisher may charge for its own service.

Runs locally over stdio

This server isn't hosted — your MCP client launches it from a package registry. Use one of the commands below, or drop the config into your client (e.g. Claude Desktop).

npmdrengrv0.10.8

Run

npx -y drengr

MCP client config

{
  "mcpServers": {
    "drengr": {
      "command": "npx",
      "args": [
        "-y",
        "drengr"
      ]
    }
  }
}

README1 min readView on GitHub

Eyes and hands for AI agents on mobile devices.

Drengr is an MCP server that gives AI agents (Claude, GPT, Gemini) the ability to see, tap, type, and navigate mobile apps — Android and iOS.

Install

bash

npm install -g drengr

Or run without installing:

bash

npx drengr

Or via shell script:

bash

curl -fsSL https://drengr.dev/install.sh | bash

Quick Setup

bash

# Check your system
drengr doctor

# Configure your MCP client (Claude Desktop, Claude Code, Cursor, etc.)
drengr setup

Add to Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

json

{
  "mcpServers": {
    "drengr": {
      "command": "drengr",
      "args": ["mcp"]
    }
  }
}

Add to Claude Code

bash

claude mcp add drengr -- drengr mcp

Add to Cursor

Add to .cursor/mcp.json in your project:

json

{
  "mcpServers": {
    "drengr": {
      "command": "drengr",
      "args": ["mcp"]
    }
  }
}

What Drengr does

3 MCP tools: drengr_look (observe screen), drengr_do (take action), drengr_query (ask questions)
Android & iOS — ADB for Android, simctl for iOS, Appium for cloud devices
Vision-first — sees your app like a human via annotated screenshots
Text mode — ~300 tokens per screen instead of 100KB images
Situation reports — after every action: what changed, what's new, is it stuck?
Screen exploration — auto-maps your app's navigation graph
Network capture — sees HTTP calls during each action