# Turbo1123/roubao

Android Automation Tool Based on Vision-Language Models

Repository: https://github.com/Turbo1123/roubao
Canonical: https://ross.abutalabs.com/products/roubao
Language: Kotlin
License: MIT
License Family: permissive
Topics: ai, ai-agents, android, android-automation, kotlin, mobile-agents, mobileagent, phone-automation
Last push: 2026-01-08T10:44:44+00:00

## Health v2 (maintenance only)
Score: 53/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 61, release rhythm 62, longevity 19
- inputs: {"age_days": 267, "days_push": 237, "days_rel": 258, "gap_med": 0, "n_releases_24m": 8}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2325, forks 243 (observed 2026-08-28T04:06:37.370066+00:00)

## What it is
Roubao is an open-source AI phone automation assistant for Android, built natively in Kotlin and powered by vision-language models. It runs entirely on-device without a PC, using Shizuku for system-level permissions and a Tools/Skills multi-agent architecture to automate tasks like ordering food or filling forms.

## Use cases
- automate android phone tasks with ai
- control my phone hands-free using a vision language model
- open source alternative to doubao phone assistant
- automate ordering food or booking on android without adb
- run mobile agent automation directly on the phone without a computer
- let ai tap and swipe through apps on my android
- batch apply to jobs automatically on my phone

## When to choose
- you want on-device android automation with no PC, ADB, or Python setup
- you need an open-source, self-hosted alternative to closed AI phone assistants
- you want to bring your own LLM/VLM API key and customize models
- you prefer a native Kotlin app over Python-based MobileAgent scripts

## When to avoid
- you need iOS or desktop automation
- you cannot or do not want to set up Shizuku for elevated permissions
- you require fully offline operation without any LLM API
- you need guaranteed reliability for critical workflows, since VLM-driven GUI automation can be error-prone

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, machine-learning, workflow-automation, computer-vision, nlp
- domain: mobile-development, artificial-intelligence
- platform: jvm
- tags: vlm, phone-automation, shizuku, mobile-agent, on-device-ai, gui-automation, no-pc-required, ai-agents, automation, android, mobile

## Member repositories
- Turbo1123/roubao (main) score 53

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:37.370066+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:38:23.089192+00:00, confidence not recorded.
  - readme: https://github.com/Turbo1123/roubao (fetched 2026-08-28T04:06:37.370066+00:00, sha c3f2b7aa79fa)
- Data as of 2026-08-30T08:39:29.467469+00:00.