OpenClaw Desktop

OpenClaw Desktop is a cross-platform desktop workspace for OpenClaw.

It turns a workflow that often starts in terminals, config files, and manual setup into a desktop product that is easier to install, understand, configure, and use every day. Whether you are an individual user, a developer, or a team running long-lived AI agents, OpenClaw Desktop gives you a unified place to manage chat, model providers, skills, automation, and agent operations.

Overview

OpenClaw is powerful, but not everyone wants to begin with CLI commands, environment variables, and hand-edited configuration.

OpenClaw Desktop focuses on these common adoption barriers:

Setup friction: download and run instead of building an environment from scratch
Configuration complexity: manage providers, agents, channels, secrets, and tools visually
Fragmented workflows: keep chat, skills, files, terminal, memory, and analytics in one workspace
Maintenance overhead: make updates, connectivity, recovery, and diagnostics part of the default experience
Low first-run success: help users reach a working result faster

Key Features

Out-of-the-box desktop experience

One-click startup without requiring a manual Node.js installation
Automatic updates
Currently ships installers for macOS and Windows
Better suited for always-available daily use, not just one-off experiments

Workspace built for real usage

Modern chat interface and multi-page desktop workspace
Unified access to chat, configuration, skills, files, terminal, and analytics
Helps turn OpenClaw from "it runs" into "it is usable every day"

Visual OpenClaw configuration

Manage AI providers, agents, channels, secrets, and tools through the UI
Reduce direct editing of config files
Make core setup more approachable for non-engineering users

Skills and extensibility

Built-in Skills page
Browse, install, import, and manage skills
Import local skills
Discover capabilities from skill directories and skill marketplaces

Automation and multi-agent workflows

Schedule recurring tasks and automated runs
Organize multi-agent workflows
Support long-running and scenario-based agent operations

Desktop-native integration

Automatic updates
Tray and system integration
Local runtime management
More reliable OpenClaw desktop experience

Typical Use Cases

Build a personal AI assistant workspace
Manage multi-agent setups and runtime status
Install and maintain skill stacks
Run scheduled tasks and automated workflows
Review files, logs, terminal output, and analytics
Use OpenClaw as a desktop control center

Screenshots

The following screenshots are taken from the v0.4.0 release article and show the actual product UI.

Chat and result rendering

File result cards

File management and preview

Provider selection

Tool configuration

Settings and connection details

Installation

Download Releases

For most users, the recommended path is to download a prebuilt installer from GitHub Releases instead of starting from source.

Download options

Currently published release packages:

Platform	Recommended package
macOS Apple Silicon	`OpenClaw.Desktop-<version>-arm64.dmg`
macOS Intel	`OpenClaw.Desktop-<version>.dmg`
Windows x64	`OpenClaw.Desktop.Setup.<version>.exe`

Notes:

Linux builds are not currently published
macOS users should prefer .dmg
Windows users should prefer .exe
.zip files are mainly useful for portable or manual extraction workflows

Start Using It

On first launch, the typical setup flow is:

Connect to or initialize OpenClaw Gateway
Configure an AI provider
Choose a default model
Install required skills
Create your first agent or start chatting immediately

Quick Start

Development

npm install
npm run bundle:node        # macOS / Linux
npm run bundle:node:win    # Windows
npm run dev

Build

npm run build:mac
npm run build:win
npm run build:linux

Build artifacts are generated in the release/ directory by default.

Compared With A Traditional OpenClaw Workflow

Scenario	Traditional workflow	OpenClaw Desktop
Setup and launch	Prepare environment and commands	Start directly from a desktop app
Provider configuration	Edit config manually	Visual management
Skill management	CLI or manual directory operations	Graphical browsing and management
Feature entry points	Split across CLI, files, and web	Unified in one desktop workspace
Day-to-day usage	Engineering-heavy	Better for regular personal and team use

CI/CD

Build workflow: build validation and artifact upload without publishing
Release workflow: publish to GitHub Releases when a v* tag is pushed

Release example:

git tag v1.0.0
git push origin v1.0.0

Documentation

English docs: docs/README.md
Chinese docs: docs/README.zh-CN.md
Build and release configuration: docs/build-release/config-build-release.md
Open-source readiness checklist: docs/open-source-checklist.md
FAQ and troubleshooting: docs/support/faq-and-troubleshooting.md
Contributing guide: CONTRIBUTING.md | Chinese: CONTRIBUTING.zh-CN.md
Security policy: SECURITY.md | Chinese: SECURITY.zh-CN.md

Acknowledgements

OpenClaw Desktop is inspired by and builds on the work of:

OpenClaw: https://github.com/openclaw/openclaw
AEGIS Desktop: https://github.com/rshodoskar-star/openclaw-desktop

License

This project is licensed under the MIT License. See LICENSE for details.