Android CLI

Android CLI

Google · Coding

Android CLI is Google's official command-line tool for Android development, built so that AI agents and developers can drive the whole toolchain from a terminal. It folds the scattered adb, sdkmanager, and Gradle commands into one `android` binary with subcommands for Android SDK management, project creation, emulator control, and app deployment. Google reports that its internal tests cut LLM token use by more than 70% and finished setup tasks 3x faster than when agents assemble those tools by hand. So who's it for? Developers running Claude Code, Codex, Gemini CLI, Cursor, or any other coding agent against an Android project.

Interface preview of Android CLI

About Android CLI

What Is Android CLI

Android CLI is the primary terminal interface for Android development. Instead of juggling Gradle, AGP, adb, and sdkmanager as separate tools, you get a single android command with structured subcommands and machine-readable output. That matters for agents: a stable command with clear parameters beats a pile of scripts that break when an environment differs.

The problem it solves is orchestration. Setting up a build environment, spinning up an emulator, and deploying a debug APK are tasks with a right answer, so they don't suit a model guessing at shell syntax. Google moved those actions into a controllable interface and paired them with Android Skills, a set of Markdown instruction files that tell an agent how to handle tricky migrations like AGP upgrades or XML-to-Compose moves. Big difference.

The limits are worth knowing. As of this writing, Android CLI is in preview, so commands and flags can shift. It also doesn't replace Android Studio. Think of it as the fast path for scaffolding and automation, with the IDE still the right home for UI tuning, profiling, and deep debugging. If you never touch a terminal, you likely won't need it.

Getting Started

  1. Install the CLI from Google's download host for your platform (Windows, macOS, or Linux), then confirm the install with android --version.
  2. Run android sdk install to pull only the SDK components your project actually needs.
  3. Create a project from an official template with android create, or point an existing one at the CLI.
  4. Start a virtual device with android emulator and deploy your build with android run.
  5. Add Android Skills with android skills add so your agent follows Google's recommended steps for tricky tasks.

Product Information

A quick look at Android CLI's pricing, supported platforms, and performance.

Free PlanYes
Paid Plans$0
PlatformWindows, macOS, Linux
DeveloperGoogle
CategoryCoding
Release DateApr 2025
Latest UpdatedJun 2025
Website Visits514.7K
Website Global RankN/A
API AvailabilityNo

Best for

The users, tasks, and scenarios where this tool fits best.

Users

  • Android developers who script builds or run CI pipelines
  • Developers using AI coding agents like Claude Code, Codex, or Gemini CLI
  • Teams onboarding new machines

Tasks

  • Project scaffolding
  • SDK and build-tool upkeep
  • Device and deployment work

Scenarios

  • Prototyping a new Android app from a prompt inside an agent
  • Migrating a codebase across framework changes
  • Automating release-maintenance chores

Key features

Unified Command Set

One android binary wraps the tasks that used to need several separate tools. You get subcommands for SDK management, project creation, device handling, and deployment, all with consistent output that scripts and agents can parse. That single entry point is the reason token use drops: an agent reads one command reference instead of probing the environment. Less guessing, fewer retries.

Agent-Friendly Output

The CLI returns structured, machine-readable results rather than walls of console noise. An agent can act on a build target or a device list without parsing human-formatted text. That's the whole point. Google built this with agent workflows in mind, and it shows in how predictable each subcommand's output is.

Project Creation From Official Templates

The android create command spins up a new project in seconds using templates Google maintains. You start from a Compose or native setup that already uses the recommended architecture, so you skip the drift that creeps in when an agent writes boilerplate from memory.

Emulator and Deployment Control

The Android emulator CLI creates and boots virtual devices, and android run deploys your app to them or to a connected phone. No IDE required. Doing device setup from the terminal removes the manual build-and-push loop that slows down an agent working on its own.

Android Skills Integration

The CLI installs Android Skills, modular instruction files in the SKILL.md format that describe how to handle tasks where LLMs tend to slip up. Google's first set covers Navigation 3, edge-to-edge layout, AGP 9 upgrades, XML-to-Compose migration, R8 rule analysis, and Play Billing updates. A skill fires when a prompt matches its metadata, so you don't have to attach docs by hand.

Knowledge Base Access

The android docs command taps a Knowledge Base drawn from Android developer docs, Firebase, and Kotlin references. Because it updates often, an agent can pull current guidance even when the model's training data is a year old. That's the difference between generated code that compiles and code that matches today's recommended patterns.

Bring-Your-Own Agent

Android CLI isn't locked to Google's own tools. It works with Gemini in Android Studio, Gemini CLI, Antigravity, and third-party agents like Claude Code and Codex. You keep your preferred agent and get the same standardized Android interface underneath. Pick your assistant, then.

Pros and cons

Pros

  • Cuts LLM token use by more than 70% on setup tasks and speeds completion to 3x, according to Google's internal experiments.
  • Replaces fragile adb, Gradle, and sdkmanager chains with one scriptable command, which makes CI jobs far less brittle.
  • Works across agent ecosystems, so teams aren't forced onto a single vendor's assistant.
  • Ships with free, regularly updated Skills and a Knowledge Base that keep output aligned with current Android guidance.
  • Installs as a single binary, with no heavy runtime to configure.

Cons

  • It's in preview, so command names and flags can change between releases and may break automation.
  • It doesn't replace Android Studio; UI work, profiling, and deep debugging still send you back to the IDE.
  • The 70% token and 3x speed figures come from Google's own tests on setup and environment tasks, not from independent benchmarks.
  • Windows installs need CMD rather than PowerShell, a rough edge for some setups.

Frequently asked questions

It's a terminal tool for Android development. You use it to install SDK components, create projects, manage emulators, and deploy apps without jumping between adb, Gradle, and sdkmanager. It's especially useful when an AI agent drives those steps.

Related content

Explore related tools, skills, and articles for Android CLI.

Android CLI Alternatives

Forefront

Forefront

Forefront · Coding

Forefront is a web platform for building with open-source AI. It lets you fine-tune leading open-source language models on your own data, evaluate how they perform, and run them through an API or export them to host yourself. Developers who want the convenience of a closed-source platform but insist on owning their models and data are the target audience here.

Free / $0 - $99/moView details
Startkit

Startkit

StartKit.AI · Coding

Startkit is a boilerplate for building AI SaaS and AI wrapper products. Think of it as an AI startup boilerplate with the boring parts already wired up: authentication, Stripe and Lemon Squeezy payments, usage limits, transactional email, and an AI API starter that talks to OpenAI, Anthropic, Groq, or Llama. You clone the repo, set your price, and start on the part of your product that people actually pay for. It's Next.js under React and Tailwind, so most of the boilerplate code already feels familiar.

Paid / $99 - $499 one-timeView details
Testim

Testim

Tricentis · Coding

Testim is an AI-powered test automation platform for building and running end-to-end tests across web, mobile, and Salesforce applications. It leans on machine learning to keep tests stable when an interface changes, so teams spend less time fixing broken selectors. Not bad for an automated testing tool you can start using today. You create tests by recording actions in a browser, then optionally add JavaScript when you need more control. It's a solid pick for busy QA teams.

Free / Custom pricing on requestView details