OpenClaw Deployment Assistant User Guide

From installation to familiarization, we help you quickly master all the features of the OpenClaw Deployment Assistant.
Whether you're a tech newbie or a developer, this tutorial makes it easy for you to get started.

📖Product introduction

Learn what the OpenClaw Deployment Assistant is and what it can do for you

OpenClaw Deployment Assistant is a cross-platform one-click deployment tool for individual users. Through automated processes and intelligent configuration wizards, we help you overcome technical barriers that may be encountered during manual deployment, and easily deploy and use OpenClaw Deployment Assistant on your own computer. 是一款面向个人用户的跨平台一键部署工具。通过自动化流程和智能配置向导，帮助您解决手动部署过程中可能遇到的技术壁垒，轻松在自己的电脑上部署和使用 OpenClaw AI 助手。

Core Features

  • One-click deployment — automatic environment detection, dependency installation, and complete deployment in 3 minutes — 自动环境检测、依赖安装，3分钟完成部署
  • Multi-model support — access to mainstream AI services such as Zhipu GLM, Tongyi Qianwen, DeepSeek, and Doubao — 接入智谱GLM、通义千问、DeepSeek、豆包等主流AI服务
  • Data privacy — fully deployed on-premise, no data uploaded to the cloud — 完全本地部署，数据不上传云端
  • Cross-platform — supports Windows, macOS, and Linux — 支持 Windows、macOS 和 Linux
  • Intelligent conversation — built-in chat interface to communicate with AI at any time — 内置聊天界面，随时与AI交流
  • Real-time monitoring — CPU, memory, and disk usage at a glance — CPU、内存、磁盘使用情况一目了然
💡

Technical architecture: Based on Rust's native development, the installation package is lightweight (tens of MB), uses very little resources, and starts up quickly.基于 Rust 原生开发，安装包轻量（几十MB），资源占用极低，启动快速。

🚀Get started quickly

Start using the OpenClaw Deployment Assistant in under 3 minutes

  1. Download and install the OpenClaw Deployment Assistant

    Go to the official download page, select the installation package suitable for your operating system, and double click to install. Supports Windows 10+, macOS 11+, and Ubuntu 20.04+.，选择适合您操作系统的安装包，双击安装即可。支持 Windows 10+、macOS 11+ 和 Ubuntu 20.04+。

  2. Launch the app and activate your membership

    When first launched, the app automatically obtains a unique device code and checks membership status. If you want to buy a membership, you can complete the payment directly within the app (supports WeChat/Alipay).

  3. Configure AI models

    On the "AI Model" page, select your AI service provider (such as Zhipu GLM), enter the API Key, and save it. For first-time users, we recommend choosing Zhipu GLM. Some models are free for individuals.。

  4. Start the service and start a conversation

    Click “Start Service” on the home control panel. After waiting for the service to be ready, click “Open Chat” to talk to the AI assistant.

✅

Congrats! Once you've completed the steps above, you have an AI personal assistant that works 24 hours a day.完成以上步骤后，您就拥有了一个 24 小时工作的 AI 私人助理。

📦Download and install

Detailed download and installation instructions

system requirements

operating system Minimum version disk space RAM
Windows Windows 10 (1809+) ≥ 500 MB ≥ 4 GB
macOS macOS 11 Big Sur ≥ 500 MB ≥ 4 GB
Linux Ubuntu 20.04 / Equivalent ≥ 500 MB ≥ 4 GB

Installation steps

Windows:

  1. Download .exe installation file 安装文件
  2. Double click to run the installer and follow the prompts to complete the installation
  3. Find the OpenClaw Deployment Assistant icon on the desktop or start menu and double-click to launch

macOS:

  1. Download the .dmg installation file 安装文件
  2. Double click Open and drag the OpenClaw Deployment Assistant into the Applications folder
  3. If you see a security prompt when running for the first time, go to “System Settings → Privacy and Security” to allow it to be opened

Linux:

  1. Download a .deb or .appImage file 或 .AppImage 文件
  2. Use dpkg -i or run AppImage directly 或直接运行 AppImage
⚠️

Note for macOS users: The first time you open it may be blocked due to security policies. Please go to “System Settings → Privacy & Security” and click “Still Open” to use it normally.首次打开可能因安全策略被阻止，请前往「系统设置 → 隐私与安全性」中点击「仍要打开」即可正常使用。

🤖AI model configuration

Select an AI service provider and configure an API key

The OpenClaw Deployment Assistant supports the configuration of multiple mainstream AI large model service providers. You only need to register your own account and obtain an API Key to access and use it. The usage fee for the AI model API interface is charged by each AI model service provider and is not related to the OpenClaw Deployment Assistant.

Supported AI service providers

🧠 OpenAI

Recommended model: GLM-5 / GLM-4-Flash：GPT-4o / GPT-4o-mini

API Address: https://dashscope.aliyuncs.com/compatible-mode/v1： https://api.openai.com/v1

Register account → | Get API Key →

☁️ Google Gemini

Recommended model: GLM-5 / GLM-4-Flash：gemini-2.0-flash

API Address: https://dashscope.aliyuncs.com/compatible-mode/v1： https://generativelanguage.googleapis.com/v1beta/openai

Register account → | Get API Key →

🎯 Anthropic Claude

Recommended model: GLM-5 / GLM-4-Flash：claude-sonnet-4

API Address: https://dashscope.aliyuncs.com/compatible-mode/v1： https://api.anthropic.com/v1

Register account → | Get API Key →

🔥 Mistral AI

Recommended model: GLM-5 / GLM-4-Flash：mistral-large / mistral-instruct

API Address: https://dashscope.aliyuncs.com/compatible-mode/v1： https://api.mistral.ai

Register account → | Get API Key →

Configuration steps

  1. Choosing an AI Service Provider

    On the "AI Model" page, click to select the AI service provider card you want to use. For the first time, we recommend using Google Gemini (free tier available for personal use).（部分模型个人免费）。

  2. Register and get an API Key

    Click “Don't have an account? Click the “Register” link. After registering on the corresponding platform, go to the API key management page to create an API Key and copy it.

  3. Fill in the configuration and save

    Paste the API Key into the “API Key” input box. The model name and API address are automatically filled in with default values, and no modification is required if there are no special requirements. Click “Verify API Connection” to confirm that the configuration is correct, then click “Save Model Configuration”.

💰

Money saving tip: Google Gemini offers a generous free tier that is suitable for everyday use. For more powerful capabilities, choose OpenAI GPT-4o or Anthropic Claude.智谱GLM平台的 GLM-4-Flash 模型对个人用户免费，适合日常使用。如需更强的模型能力，可选择 GLM-5 或其他付费模型。

🔑

API Key security: Your API key will be stored encrypted locally and will not be uploaded to any server. Don't share your API Key with others.您的 API Key 将被加密存储在本地，不会上传到任何服务器。请勿将 API Key 分享给他人。

🔑How to obtain an API key

Take OpenAI as an example to teach you how to register an account and obtain an API Key

If you don't have an AI model platform account and API key yet, don't worry. The following demonstrates the general process of registering and obtaining an API Key. The steps are similar across all providers.平台为例，详细演示注册并获取 API Key 的完整流程。

💡

Which provider to choose? Google Gemini offers a free tier ideal for beginners. OpenAI GPT-4o provides industry-leading quality. DeepSeek offers competitive pricing. Choose based on your needs.智谱AI 提供部分免费模型（如 GLM-4-Flash），注册简单，非常适合首次使用的用户。

  1. Open the AI provider's website

    Visit your chosen AI provider's website (e.g. OpenAI Platform), find and click the "Sign Up" or "Login" button. 打开官网页面，在网页右上角找到并点击 「登录/注册」 按钮。

    点击登录/注册按钮 alt="Click Login/Register button">

  2. Register/Login

    Register an account using your email address. Follow the prompts to verify your email and complete the registration process. 按钮。手机收到短信验证码后，输入到验证码编辑框中，勾选同意用户协议和隐私条款，然后点击 「登录/注册」。

    注册/登录窗口 alt="Registration/Login window">

  3. Go to the API Key management page

    After logging in, navigate to the API Key management page (usually found in Settings or your account menu). Click to create a new API key.，鼠标点击头像，在弹出的菜单中点击 「API Key」 菜单项。

    选择 API Key 菜单 alt="Select API Key menu">

  4. Adding a new API Key

    After entering the API Key management page, click the “Add New API Key” button in the upper right corner of the page. 按钮。

    API Key管理页面 - 点击添加按钮 alt="API Key management page - click Add button">

  5. Enter an API Key name

    In the New API Key window that pops up, enter a name in the edit box (for example: openclaw) and click the “OK” button.），然后点击 「确定」 按钮。

    新建 API Key - 输入名称 alt="Create new API Key - enter name">

  6. Copy API Key

    Once created, an API key called openclaw will be added to the list. Click the “Copy” button next to copy the API Key to your clipboard. 按钮，将 API Key 复制到剪贴板。

    复制 API Key alt="Copy API Key">

  7. Configure an API Key in OpenClaw

    Open the OpenClaw Deployment Assistant, go to the AI model configuration page, paste the API Key you just copied in the API Key edit box, and the API Key settings for the AI model are complete! 编辑框中粘贴刚才复制的 API Key，到此 AI 模型的 API Key 设置就完成了！

    Configuration interface in the deployment wizard:

    部署向导 - 配置 API Key alt="Deployment wizard - configure API Key">

    AI model management page in the dashboard:

    控制面板 - AI 模型管理 alt="Control panel - AI model management">
✅

Congrats! Once the API Key is configured, you can start the Gateway service and start using the AI Assistant for intelligent conversations.API Key 配置完成后，您就可以启动 Gateway 服务，开始使用 AI 助手进行智能对话了。

⚠️

Security Reminder: Please keep your API Key safe and do not share it with others or post it on public platforms. If the API Key is leaked, please go to your AI provider's platform to delete and re-create it as soon as possible.请妥善保管您的 API Key，不要将其分享给他人或发布到公开平台。如果 API Key 泄露，请及时前往智谱AI平台删除并重新创建。

⚡Service management

Start, stop, and restart the OpenCLAW Gateway service

The OpenClaw Deployment Assistant forwards AI requests through the local Gateway service. Before using the chat feature, you need to start the Gateway service.

Basic operations

Operation Description Usage scenarios
Start service Start OpenClaw Gateway First use or after service stops
Stop service Close Gateway to release resources Save resources when not in use
Restart service Restart after stopping Restart needed to apply changes after configuration

How to operate

  1. Go to the “Home” control panel
  2. Check the service control card on the top left
  3. If the service is not running, click the “Start Service” button按钮
  4. When the service is running, you can click “Stop Service” or “Restart Service” in the lower right corner或右下角「重启服务」
  5. Status indicators: 🟢 Running | 🟡 Testing | ⚫ Stopped
ℹ️

Auto Detect: The app automatically detects the Gateway service status every 10 seconds without a manual refresh.应用会每 10 秒自动检测一次 Gateway 服务状态，无需手动刷新。

💬Smart conversation

Communicate in natural language with AI assistants

Once the Gateway service has started, click the “Open Chat” button on the home page to begin a conversation with the AI assistant.按钮即可开始与 AI 助手对话。

How to use

  1. Confirm that the home page status bar shows “OpenClaw is running”
  2. Click the “Open Chat” button to open a dedicated chat window
  3. Type your question or instructions in the input box
  4. The AI assistant will generate answers in real time

Conversation Skills

  • Be as clear and specific as possible about your needs地描述您的需求
  • Provide enough contextual information to get more accurate answers以获得更准确的回答
  • Multiple rounds of conversation are possible, and the AI will remember the context of the conversation，AI 会记住对话上下文
  • Try different ways of formulating questions to find the best interaction
🚨

If “Service not started” appears when you click “Open Chat”, please start the Gateway service on the home page first.

📊System monitoring

View system resource usage in real time

The bottom of the homepage shows a real-time system resource monitoring panel to help you understand the computer's operating status:

Monitoring item View system resource usage in real time Refresh frequency
CPU Usage Current CPU utilization percentage Every 5s
Memory usage Used memory / Total memory Every 5s
Disk Space Used disk / Total disk Every 5s
Gateway Address Local Gateway access URL Real-time
Runtime Runtime since this startup Every 1s
Version Info Installed OpenClaw version At startup
📈

Click the “Refresh” button in the upper right corner of the System Resources panel to update the data manually.

⚙️Configuration

Customize the operating parameters of the OpenClaw Deployment Assistant

Click “Configuration Management” in the left navigation to enter the configuration page. You can adjust the following settings:

Basic configuration

  • Installation path: OpenClaw installation directoryOpenClaw 的安装目录
  • Gateway port: local service listening port, default 18789本地服务监听端口，默认 18789
  • Log level: Debug/Info/Warning/ErrorDebug / Info / Warning / Error

Advanced Configurations

  • Auto start: Automatically start the OpenClaw Deployment Assistant when booting开机时自动启动 OpenClaw Deployment Assistant
  • Debug mode: Record detailed logs after activation to facilitate troubleshooting启用后记录详细日志，方便排查问题
  • Automatic updates: Detects and installs the latest version检测并安装最新版本
⚠️

After modifying the port, you need to restart the Gateway service to take effect. You can click “Restart Service” on the homepage.需要重启 Gateway 服务才能生效。可以在首页点击「重启服务」。

📝Log Management

View and manage system operation logs

In the “Log Center”, you can view all system operation records and operation logs.

Function description

  • Level filtering: Filter logs by four levels: information, success, warning, and error按信息、成功、警告、错误四个级别过滤日志
  • Keyword search: enter keywords to quickly locate specific logs输入关键词快速定位特定日志
  • Clear log: Clear all history log records with one click一键清除所有历史日志记录

Log level description

Level Color Description
INFO 🔵 Blue General information, such as operation records
SUCCESS 🟢 Green Operation successful, such as service start successful
WARNING 🟠 Orange Warning message, needs attention
ERROR 🔴 Red Error message, needs processing

🔧One-click deployment

Automate the deployment of OpenClaw to your computer

The deployment wizard will guide you through the automated deployment of OpenCLAW and address technical barriers that may be encountered during manual deployment. The entire process is divided into 6 steps:

  1. Environment Check

    Automatically detects the operating system version, disk space, internet connection, and necessary dependencies (Python, Git, etc.). If it's missing, an automatic installation will be attempted.

  2. Select a working directory

    Set the OpenCLAW installation path. The default is the user directory. Make sure the directory you select has enough disk space.

  3. Choosing an AI Service

    Select the AI large model service you want to use (OpenAI/Google Gemini/DeepSeek/Anthropic Claude, etc.).

  4. Configure AI parameters

    Fill in the configuration information such as the API key, model name, and API address, and click “Verify” to test the connection.

  5. Download and deploy

    Download the OpenCLAW core program through the acceleration channel, unzip it, and execute the initialization. This step requires an internet connection, and progress is displayed in real time.

  6. Validate and complete

    Automatically verify that the service is running properly and that the API connection is working properly. Once verified, you can start using it.

⏱️

Deployment time: Full deployment usually takes 3-10 minutes, depending on network speed and system performance.完整部署通常需要 3-10 分钟，取决于网络速度和系统性能。

👑Member services

Learn about membership benefits and how to buy

Member Level

Level Function Description
Free users Can browse basic pages Cannot use installation and deployment features
VIP Member Full Installation, Deployment, and Usage Single User, Single Machine License, Full Functionality Unlocked

How to buy

  1. Click the hamburger menu on the right side of the title bar ☰ → “Open a member” → 「开通会员」
  2. Choose the membership package that's right for you
  3. Use WeChat or Alipay to scan the QR code to pay
  4. Automatically activated after successful payment, no additional action required
💡

No login required: Payment memberships are based on a device code, and can be purchased and used without logging in to an account. If you need to manage across devices, you can choose to log in and bind.支付会员基于设备码，无需登录账号即可购买和使用。如需跨设备管理，可选择登录并绑定。

❓FAQs

Quickly find solutions to your problems

On the "AI Model" configuration page in OpenClaw, select an AI service provider and the corresponding registration and API Key links will be displayed. Click the link to register on the platform, then create an API Key on the key management page. We recommend Google Gemini for beginners — it offers a free tier for personal use.

Please troubleshoot with the following steps:

  1. Check if the port is occupied by another program (default 18789)
  2. Confirm OpenClaw is properly installed (try redeploying)
  3. Check the "Log Center" for detailed error messages
  4. If the problem persists, contact customer service

  1. 检查端口是否被其他程序占用（默认 18789）
  2. 确认 OpenClaw 已正确安装（可尝试重新部署）
  3. 查看「日志中心」获取详细错误信息
  4. 如果问题持续，请联系在线客服
On the "AI Model" page, re-select the provider, enter the new API Key and model name, and save. The system will automatically restart the Gateway service and the new configuration will take effect immediately.
This means your API Key is valid but the corresponding AI service account has insufficient balance. Please top up on the respective platform. If using Google Gemini, you can use the free tier quota. Alternatively, switch to DeepSeek for competitive pricing.
This is caused by macOS security policies. Go to "System Settings → Privacy & Security", find the blocked app prompt at the bottom, and click "Open Anyway".
All data (including configurations, chat history, API Keys) is stored on your local computer and will not be uploaded to the cloud. API Keys are stored with encryption to fully protect your data security and privacy.
Membership is bound to the device code by default (single user, single device license). For cross-device management, log in to your account and bind your membership to it, then you can manage across different devices.
Click the hamburger menu ☰ on the right side of the title bar → "Redeploy" to re-enter the deployment wizard. Your existing configuration will be preserved in case you want to modify. → 「重新部署」，将重新进入部署向导流程。原有配置会保留，您可以根据需要修改。

🎧Contact customer service

NEED HELP? Contact us anytime

Email Support

  • Email：support@pcgogo.com
  • Working hours:Official working days 9:00-11:50 / 13:30-18:00 | Official holidays 10:00-22:00
  • Friendly reminder:Please leave a message during non-business hours, and customer service will reply as soon as possible

📧 Contact customer service now

© 2026 TECHVISTA Co., Ltd. All rights reserved. | Back to official website | Terms of Service | Email Support | Member Service Agreement | Online Customer Service