AAswordman

Operit: Run AI Agents and Chat on Android

Operit is an Android AI agent platform that connects cloud or local models to device tools, terminals, browsers, and project workspaces. It suits Android users who want AI-assisted automation, development, and chat in one app.

KotlinLGPL-3.0AndroidAIAI Agents
View on GitHub Compare
Stars
8.3k
Forks
690
Stars, last 30 days
+880

Our take

Health score

71/100

How it is scored
Activity
78
Community
48
Issues
83
Pull requests
75

+880 stars in the last 30 days

  • Merges more pull requests than 78% of the projects we track

Project history

Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.

Selected snapshot

Stars

9,656

Forks

845

Change since Nov 16, 2026: +28 stars·+2 forks

Overview

Operit is an AI chat and agent platform for Android. It connects configurable cloud and local models to Android capabilities, files, a browser, terminals, and development workspaces so users can ask an agent to carry out practical tasks rather than only answer questions.

It is aimed at people who want an extensible, phone-based environment for AI-assisted automation and development. Operit is the Android edition; its README points to Operit 2 as a separate cross-platform successor.

Key Features

  • Connect OpenAI-compatible, Anthropic, Gemini, and other model services, or run supported MNN and GGUF local models.
  • Use tools for file handling, networking, search, media, system operations, and software development, with configurable permission policies.
  • Automate Android interfaces through Accessibility, Shizuku, or Root, subject to device support and permissions.
  • Work with project files and code in workspaces, preview web projects, and run an Ubuntu user space and terminal.
  • Use a built-in browser and browser agent for page interaction and inspection.
  • Extend capabilities with ToolPkg, MCP services, Skills, and visual workflows.
  • Keep conversation and memory data locally, with options for long-term memory and character-based chats.

Use Cases

  • Android power users can automate device tasks when they are willing to grant and manage the required system permissions.
  • Developers can edit and test small projects on a phone, use terminal tools, and ask an agent to work with workspace files.
  • Users who want model choice can connect their own cloud endpoints or try supported local inference, depending on device capacity.
  • People managing repeatable routines can compose tools into workflows and trigger them manually, on schedules, or through integrations.

Getting Started

Install the latest APK from the GitHub Releases page, then follow the in-app setup to configure models and permissions. See the README and user guide for setup details and usage guidance.

Alternatives

  • UFO: UFO targets Windows desktop and multi-device automation, while Operit is an Android app centered on on-device tools, terminals, browsers, and workspaces.
ProjectLanguageLicenseStarsStatus
OperitKotlinLGPL-3.08.3kNot checked yet
UFOPythonMIT9.9kActive

Considerations

  • The README specifies Android 8.0 or newer and ARM64 devices.
  • Device automation depends on authorization and may require Accessibility, Shizuku, or Root; some capabilities also depend on device support.
  • Local model and terminal usage can require substantial storage and memory. Resource use varies with installed models and tools.
  • Cloud-model requests go to the provider endpoint configured by the user. Marketplace and related integrations may also contact third-party services.
  • Web Chat and the HTTP API are disabled by default. Review authentication and network exposure before enabling them.
  • The main code uses LGPL-3.0, while tools, examples, templates, and dependencies may have different licenses.

Found this useful?

Share it with someone who would like Operit.

Comparisons