🤖 OpenClaw on Android (OCA)
🚀 Turn any Android phone into a 24/7 AI server — one command, zero hassle.
Native performance via Termux & glibc. No Proot. No VM. Pure ARM64 power.
🚀 Quick Start • ⚡ Features • 🦙 Local LLM • 📖 Docs • 🏗 Architecture • 🤝 Community
🌟 The Vision
Your old Android phone is not e-waste. It's a powerful ARM64 server waiting to happen.
OCA seamlessly installs the OpenClaw AI ecosystem directly onto your device via Termux. This completely bypasses sluggish Linux distributions (like Ubuntu on Proot), running natively with full glibc compatibility instead of Android's default Bionic libraries.
| Before OCA ❌ | After OCA ✅ |
|---|---|
| Slow Proot containers | Native ARM64 execution |
| Bionic libc limitations | Full glibc compatibility |
| Manual setup (hours) | Zero-Prompt Auto-Install (2 mins) |
| Limited AI tools | Optimized Qwen/Gemini Arsenal |
| Individual tool prompts | Total Autonomous Deployment |
| Static installation | Auto-updating ecosystem |
🚀 Quick Start
Deploying OCA takes under 5 minutes
Step 1: Prepare Your Phone ⚙️
Configure Developer Options, Stay Awake, and battery optimization to prevent Android from killing Termux.
👉 Read Phone Setup Guide for step-by-step instructions.
Step 2: Install Termux 📱
⚠️ Important: The Play Store version of Termux is discontinued. You must install from F-Droid.
- Open browser and go to f-droid.org
- Search for Termux → Tap Download APK
- Install and allow "Install from unknown sources"
Step 3: Initial Setup 🔧
Open Termux and run:
pkg update -y && pkg install -y curl
Step 4: Install OCA 🚀
curl -sL https://raw.githubusercontent.com/PsProsen-Dev/OpenClaw-On-Android/master/bootstrap.sh | bash && source ~/.bashrc
Takes 3-10 minutes depending on network and device.
Step 5: Automatic glibc extras layer
On the OpenClaw platform, OCA now installs Homebrew (Linuxbrew) and the Go toolchain automatically once the glibc runtime is ready.
Why automatic?
glibc-runnerprovides the glibc dynamic linker that standard Linux binaries need on Android. Since Homebrew and many Go-based helper tools depend on that same Linux-compatible layer, OCA treats them as part of the default glibc stack instead of separate opt-in extras.
After install completes, verify the extra toolchain:
source ~/.bashrc && brew --version && go version
Step 6: Start OpenClaw Setup
openclaw onboard
Follow the on-screen instructions.
Step 7: Start Gateway
openclaw gateway
⚠️ Important: Run directly in Termux app, not via SSH. SSH session disconnect will stop the gateway.
That's it. You now have a full AI server on your phone.
⚡ Features
| 📱 Native Execution | ☁️ Full Node.js v24 |
|---|---|
| OpenClaw runs bare-metal in Termux | glibc-patched Node.js v24.14.1 |
| No Proot overhead | Official linux-arm64 binaries |
| 100% ARM64 optimized | Bypasses Android linker restrictions |
| 🤖 AI CLI Tools | 🦙 Local LLM |
|---|---|
| Qwen Code, Gemini CLI, OpenCode | node-llama-cpp + Ollama support |
| Zero-prompt auto-setup | Run models locally (experimental) |
| Cloud API routing | ☁️ NEW: Ollama Cloud Models |
| 🌐 Remote Access | 🛡️ Safe Root |
|---|---|
| SSH server on port 8022 | oca-root wrapper for rooted devices |
| Remote terminal access | Selective root commands |
| 24/7 headless operation | No system compromise |
| 🔄 Auto-Updates | 🔧 Master CLI |
|---|---|
One-command updates via oca --update |
oca command for all operations |
| Automatic security patches | Global status, health, & maintenance |
| Rolling release model | Platform-aware architecture |
🔧 CLI Reference
After installation, use the oca command for managing your installation:
| Command | Description |
|---|---|
oca --update |
Update OpenClaw and Android patches |
oca --status |
Show global installation health |
oca --clean |
Purge caches and binary junk |
oca --uninstall |
Clean removal of OCA ecosystem |
oca --version |
Show version info |
oca --help |
Show Jarvis-grade help |
Update example:
oca --update && source ~/.bashrc
This updates: OpenClaw core, code-server, OpenCode, AI CLI tools, Android patches
🦙 Local LLM
Run AI models directly on your phone!
☁️ NEW: Ollama Cloud Models - No local resources needed!
OCA now supports local LLM inference via node-llama-cpp and Ollama (including cloud models!).
☁️ Ollama Cloud Models (Recommended)
Run powerful models in the cloud — zero local RAM/storage usage!
# Pull and launch with cloud model
ollama pull kimi-k2.5:cloud
ollama launch openclaw --model kimi-k2.5:cloud
Recommended Cloud Models:
kimi-k2.5:cloud- Multimodal reasoning (64k context)minimax-m2.5:cloud- Fast coding (64k context)glm-5:cloud- Reasoning & code generationgpt-oss:120b-cloud- High-performance (128k context)
🏠 Local Models (Experimental)
⚠️ Important Constraints (click to expand)
| Constraint | Requirement | Reality Check |
|---|---|---|
| RAM | 2-4GB free | Phone RAM is shared with Android |
| Storage | 4-70GB+ | Model files are large |
| Speed | CPU-only | No GPU offloading on Android |
| Use Case | Testing/Experimentation | Cloud APIs for production |
Quick Start
# Option 1: node-llama-cpp (Recommended)
npm install -g node-llama-cpp --ignore-scripts
# Option 2: Ollama (Full Server)
curl -fsSL https://ollama.com/install.sh | sh
Model Recommendations
| Model | Size | RAM Needed | Speed | Best For |
|---|---|---|---|---|
| TinyLlama 1.1B | ~670MB | 2GB | ⚡⚡⚡ | Testing |
| Phi-3 Mini | ~2.3GB | 4GB | ⚡⚡ | Light tasks |
| Llama 3.2 1B | ~670MB | 2GB | ⚡⚡⚡ | Mobile-friendly |
| Mistral 7B | ~4.1GB | 8GB | ⚡ | Advanced only |
📖 Read Full Local LLM Guide for detailed setup, troubleshooting, and cloud comparison.
🏗 Project Architecture
How OCA Works
┌─────────────────────────────────────────────────────────┐
│ Android Device (Termux) │
│ │ │
│ ▼ │
│ ┌──────────────────┐ │
│ │ glibc-runner │ │
│ │ (ld.so wrapper) │ │
│ └────────┬─────────┘ │
│ │ │
│ ▼ │
│ ┌──────────────────┐ │
│ │ Node.js v24 │ │
│ │ linux-arm64 │ │
│ └────────┬─────────┘ │
│ │ │
│ ┌─────────────┴──────────────┐ │
│ ▼ ▼ │
│ ┌──────────────┐ ┌──────────────┐ │
│ │ OpenClaw │ │ Local LLM │ │
│ │ Gateway │ │ (Optional) │ │
│ │ │ │ │ │
│ │ • AI CLIs │ │ • llama.cpp │ │
│ │ • SSH │ │ • Ollama │ │
│ │ • clawdhub │ │ • GGUF │ │
│ └──────────────┘ └──────────────┘ │
└─────────────────────────────────────────────────────────┘
Technical Highlights
- Pacman
glibc-runner: Injectsld-linux-aarch64.so.1to bypass Android's restricted linker - Path Rewriting: UNIX paths (
/tmp,/bin/sh) dynamically mapped to Termux prefixes - JS Runtime Shims:
glibc-compat.jspolyfillsos.cpus()andos.networkInterfaces()for V8
📖 Documentation
Everything you need to run OCA like a pro
| 📚 Guide | Description |
|---|---|
| 🚀 Quick Start | Get running in 5 minutes |
| 📱 Phone Setup | Developer Options, Stay Awake, Battery |
| 🔧 Installation | Full 8-step installer breakdown |
| 🤖 AI CLI Tools | Qwen, Claude, Gemini, Codex setup |
| 🦙 Local LLM | Run models locally (node-llama-cpp, Ollama) |
| 🌐 Dashboard Connect | Multi-device management from PC |
| 🔐 SSH Setup | Remote access configuration |
| ⚙️ Configuration | Manage settings and preferences |
| 🔧 Troubleshooting | Common errors and fixes |
| 👻 Phantom Process Killer | Android 12+ fix |
📑 Browse all docs:
docs/
🛑 Important: Android 12+ Users
If you see [Process completed (signal 9)], Android's Phantom Process Killer has terminated Termux.
Fix it in 30 seconds:
adb shell settings put global development_settings_enabled 1
adb shell settings put global max_phantom_processes 64
📱 Phone Preparation
Essential for 24/7 Operation
Android may kill background processes or throttle them when the screen is off. For 24/7 operation:
| Setting | Purpose | How To |
|---|---|---|
| Developer Options | Enable advanced controls | Settings → About → Tap Build Number 7x |
| Stay Awake | Prevent CPU throttling | Developer Options → Stay Awake |
| Battery Optimization | Prevent app killing | Settings → Apps → Termux → Battery → Unrestricted |
| Charge Limit | Protect battery during 24/7 use | Use AccuBattery or similar |
📖 Read Complete Phone Setup Guide for detailed instructions.
🌐 Dashboard & Remote Access
SSH Setup
Access your OCA dashboard from PC browser via SSH tunnel:
# From your PC terminal
ssh -L 3000:localhost:3000 -L 8080:localhost:8080 [email protected] -p 8022
📖 Read SSH Setup Guide for complete instructions.
Dashboard Connect (Multi-Device)
Running OCA on multiple devices? Use Dashboard Connect to manage them from your PC:
- Save connection settings (IP, token, ports) for each device with a nickname
- Generates SSH tunnel commands automatically
- Switch between devices with one click
- Your data stays local — Settings saved only in browser localStorage
💡 Tip: Name your devices (e.g., "Old Pixel", "Bedroom Phone") for easy identification.
🏗 Platform Architecture (Developer Info)
RTX⚡1/RTX⚡2/RTX⚡3 Dependency Layers
OCA uses a platform-plugin architecture that separates platform-agnostic infrastructure from platform-specific code:
┌─────────────────────────────────────────────────────────────┐
│ Orchestrators (install.sh, update-core.sh, uninstall.sh) │
│ ── Platform-agnostic. Read config.env and delegate. │
├─────────────────────────────────────────────────────────────┤
│ Shared Scripts (scripts/) │
│ ── RTX⚡1: install-infra-deps.sh (always) │
│ ── RTX⚡2: install-glibc.sh, install-nodejs.sh (conditional)│
│ ── RTX⚡3: Optional tools (user-selected) │
├─────────────────────────────────────────────────────────────┤
│ Platform Plugins (platforms/<platform>/) │
│ ── config.env: declares dependencies │
│ ── install.sh / update.sh / uninstall.sh / ... │
└─────────────────────────────────────────────────────────────┘
Dependency layers:
| Layer | Scope | Examples | Controlled by |
|---|---|---|---|
| RTX⚡1 | Infrastructure (always) | git, pkg update | Orchestrator |
| RTX⚡2 | Platform runtime (conditional) | glibc, Node.js, build tools | config.env flags |
| RTX⚡3 | Optional tools (user-selected) | tmux, code-server, AI CLIs | User prompts |
Installed Components
Core Infrastructure (RTX⚡1): git
Platform Runtime (RTX⚡2): pacman, glibc-runner, Node.js v24, python, make, cmake, clang, binutils
OpenClaw Platform: OpenClaw, clawdhub, PyYAML, libvips
Optional Tools (RTX⚡3): tmux, ttyd, dufs, android-tools, code-server, OpenCode, Qwen Code CLI, Gemini CLI.
🎯 What's New in v01.04.2026
| Feature | Description |
|---|---|
| RTX⚡ Protocol | New tiered 3-tier architecture (Infra, Engine, Arsenal) |
Master CLI (oca) |
Updated for protocol-aware health & sync |
| Playwright Arsenal | Native Chromium FHS automation suite |
| Repo Cleanup | Optimized infrastructure for 0% crash rate |
🤝 Community
Join the discussion! Ask questions, share your Android setup, or request features.
🙏 Credits & License
🚀 Ready to transform your Android phone?
curl -sL https://raw.githubusercontent.com/PsProsen-Dev/OpenClaw-On-Android/master/bootstrap.sh | bash
Your phone is now an AI server. ⚡

