An expert AI agent skill designed for AOSP compilation, HAL analysis, and Android Framework customization.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install android-dev-agent
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install android-dev-agent using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Android Dev Agent is a specialized technical module within the Openclaw Skills ecosystem designed for engineers working on deep-level Android development. It excels at handling complex tasks such as AOSP source code compilation, Hardware Abstraction Layer (HAL) transitions, and custom ROM development. By leveraging high-reasoning models like Claude 3.5 Sonnet or Opus, it provides precise code modifications and architectural insights.
This skill ensures a clean development environment by utilizing isolated worktrees and automated reporting. Whether you are migrating from HIDL to AIDL or troubleshooting kernel configurations, this tool streamlines the high-stakes workflow of system-level Android engineering.
To integrate this capability into your Openclaw Skills environment, ensure your workspace has access to the Android build tools and repository sync capabilities:
# Initialize the AOSP repository
repo init -u https://android.googlesource.com/platform/manifest -b android-14.0.0_r1
repo sync -c -j8
# Prepare the build environment
source build/envsetup.sh
lunch aosp_x86_64-eng
The skill maintains a strict organizational structure for all generated technical artifacts:
| Data Type | Path / Format | Description |
|---|---|---|
| Reports | memory/reports/android-dev-*.md |
Final analysis and task summaries. |
| Workspaces | memory/worktrees/android-{id}/ |
Isolated directories for builds and patches. |
| Build Logs | Standard Output / Redirected Logs | Captured logs from m, mm, and mmm commands. |
| Patches | .patch or Inline Markdown |
Source code modifications and workarounds. |
Loading
A gamification skill that transforms real-world interactions into cultivation progress within an immersive fantasy RPG framework.

An enterprise-grade execution engine for running multiple AI agent tasks concurrently with robust error isolation and timeout management.

CryptoLint is a powerful security scanner designed to detect cryptographic misuse, deprecated algorithms, and hardcoded secrets within your codebase.

A streamlined web dashboard for managing AI agent email accounts through Gmail and Gog CLI integrations.

A comprehensive security framework for AI agents that implements multi-level trust ladders, prompt injection defenses, and strict approval workflows.

A comprehensive automation toolkit to build, launch, and manage an AI-operated business in just 72 hours.








































