Drengr

Eyes and hands for AI agents on Android and iOS devices, enabling screen observation, interaction, and autonomous navigation via a single Rust binary.

Drengr gives AI agents the ability to see, understand, and interact with Android and iOS mobile devices. Built as a single Rust binary with three core MCP tools (observe screen, take action, query elements), it supports both vision-based and text-only modes for efficient mobile automation. The platform includes a situation engine that tracks screen changes, crash detection, and stuck states between actions. Drengr works with any MCP client through stdio transport and supports local devices via ADB/simctl as well as cloud device services.