# minitap-ai/mobile-use

AI agents can now use real Android and iOS apps, just like a human.

Repository: https://github.com/minitap-ai/mobile-use
Canonical: https://ross.abutalabs.com/products/mobile-use
Homepage: https://minitap.ai
Language: Python
License: Apache-2.0
License Family: permissive
Topics: agents, ai, browser-use, langgraph, mobile, mobile-use, python, qa, langchain, langgraph-python
Last push: 2026-08-26T13:33:07+00:00

## Health v2 (maintenance only)
Score: 73/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 65, longevity 27
- inputs: {"age_days": 382, "days_push": 7, "days_rel": 233, "gap_med": 9, "n_releases_24m": 10}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2784, forks 237 (observed 2026-08-28T04:07:21.840327+00:00)

## What it is
An open-source Python AI agent framework that controls real Android and iOS devices using natural language commands, built on LangGraph. It navigates app UIs, performs tasks, and scrapes structured data from mobile apps, and also powers a commercial autonomous mobile QA product.

## Use cases
- automate tasks on my android phone with natural language
- control an ios app with an ai agent
- scrape structured data from a mobile app
- run autonomous e2e tests on mobile apps
- let an llm agent navigate app uis like a human
- build a phone automation agent with langgraph
- regression test mobile apps without writing scripts

## When to choose
- you want natural-language automation of real Android or iOS devices
- you need UI-aware agents that use accessibility tree data
- you want to extract structured data from mobile apps
- you need an open-source alternative to scripted mobile testing tools like Appium or Maestro

## When to avoid
- you need to automate mobile games, which lack accessibility tree data
- you need a fully managed QA service rather than a self-hosted agent library
- you need stable, production-hardened tooling - the project is still quickly evolving

## Facets
- artifact type: library
- maturity: active
- function: agent-framework, workflow-automation, testing, e2e-testing, web-scraping, llm-inference, mcp
- domain: mobile-development, testing, large-language-models
- platform: python, cross-platform, cli
- tags: mobile-automation, natural-language-control, ui-automation, androidworld-benchmark, langgraph, device-control, qa-automation, ai-agents, automation, android, ios

## Member repositories
- minitap-ai/mobile-use (main) score 73

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:21.840327+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 2026-08-30T08:16:03.076782+00:00, confidence not recorded.
  - readme: https://github.com/minitap-ai/mobile-use (fetched 2026-08-28T04:07:21.840327+00:00, sha ce2d2498efed)
  - homepage: https://minitap.ai (fetched 2026-08-29T09:55:24.135450+00:00, sha db766b5eb7d4)
  - site_page: https://www.minitap.ai/docs (fetched 2026-08-29T09:55:24.146334+00:00, sha e845159a1b26)
  - site_page: https://www.minitap.ai/about (fetched 2026-08-29T09:55:24.147895+00:00, sha 707c6f157da5)
  - site_page: https://www.minitap.ai/pricing (fetched 2026-08-29T09:55:24.144475+00:00, sha 18785b63369e)
- Data as of 2026-08-30T08:39:29.467469+00:00.