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 workflows and intelligent setup wizards, it helps you overcome the technical barriers of manual deployment and quickly run OpenClaw on your own computer.

Core Features

  • One-click deployment — automatic environment detection, dependency installation, and complete deployment in about 3 minutes
  • Multi-model support — connect to mainstream AI services such as Zhipu GLM, Tongyi Qianwen, DeepSeek, and Doubao
  • Data privacy — fully local deployment with no data uploaded to the cloud
  • Cross-platform — supports Windows, macOS, and Linux
  • Intelligent conversation — built-in chat interface so you can talk with AI anytime
  • Real-time monitoring — CPU, memory, and disk usage at a glance
💡

Technical architecture: Built with native Rust, the installer stays lightweight, uses minimal resources, and starts quickly.

🚀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, choose the installer for your operating system, and double-click to install. Supports Windows 10+, macOS 11+, and 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, enter the API Key, and save it. For first-time users, we recommend starting with Zhipu GLM because some models offer free individual quotas.

  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 finish the steps above, your AI personal assistant is ready to work for you around the clock.

📦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.

🤖AI model configuration

Select an AI service provider and configure an API key

The OpenClaw Deployment Assistant supports the configuration of multiple mainstream AI model providers. You only need to register your own account and obtain an API Key to access and use them. Fees for AI model APIs are charged by each provider and are not related to the OpenClaw Deployment Assistant itself.

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 the AI service provider card you want to use. For first-time users, we recommend Google Gemini because it offers a free tier for personal use.

  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 for everyday use. If you need more powerful capabilities, choose OpenAI GPT-4o or Anthropic Claude.

🔑

API Key security: Your API key will be stored encrypted locally and will not be uploaded to any server. Do not share your API Key with others.

🔑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 or API Key yet, don't worry. The following walkthrough demonstrates the general process for registering and obtaining an API Key. The steps are similar across providers.

💡

Which provider should you choose? Google Gemini offers a free tier that is ideal for beginners. OpenAI GPT-4o provides industry-leading quality, and DeepSeek offers competitive pricing. Choose according to your needs.

  1. Open the AI provider's website

    Visit your chosen AI provider's website (for example, OpenAI Platform) and click the "Sign Up" or "Login" button.

    Click Login/Register button

  2. Register/Login

    Register an account with your email address. Follow the prompts to verify your email and complete the registration process.

    Registration/Login window

  3. Go to the API Key management page

    After logging in, go to the API Key management page (usually under Settings or your account menu) and create a new API Key.

    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.

    API Key management page - click Add button

  5. Enter an API Key name

    In the new API Key dialog, enter a name in the input box (for example: openclaw) and click the “OK” button.

    Create new API Key - enter name

  6. Copy API Key

    Once created, an API Key named openclaw will appear in the list. Click the “Copy” button next to it to copy the API Key to your clipboard.

    Copy API Key

  7. Configure an API Key in OpenClaw

    Open the OpenClaw Deployment Assistant, go to the AI model configuration page, and paste the API Key you just copied into the API Key input box. After saving, the AI model API Key setup is complete.

    Configuration interface in the deployment wizard:

    Deployment wizard - configure API Key

    AI model management page in the dashboard:

    Control panel - AI model management
✅

Congrats! Once the API Key is configured, you can start the Gateway service and begin using the AI assistant for intelligent conversations.

⚠️

Security reminder: Keep your API Key safe. Do not share it with others or post it on public platforms. If the API Key is leaked, delete it and create a new one on your provider's platform as soon as possible.

⚡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.
  4. When the service is running, click “Stop Service” or “Restart Service” in the lower-right corner.
  5. Status indicators: 🟢 Running | 🟡 Testing | ⚫ Stopped
ℹ️

Auto Detect: The app automatically checks the Gateway service status every 10 seconds, so no manual refresh is required.

💬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.

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

  • Describe your needs as clearly and specifically as possible.
  • Provide enough context to get more accurate answers.
  • Use multi-turn conversations; the AI will remember the context.
  • 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 directory
  • Gateway port: local service listening port, default 18789
  • Log level: Debug/Info/Warning/Error

Advanced Configurations

  • Auto start: automatically launch the OpenClaw Deployment Assistant when the system boots.
  • Debug mode: record detailed logs after activation to simplify troubleshooting.
  • Automatic updates: detect and install the latest version.
⚠️

After modifying the port, you need to restart the Gateway service for the change to take effect. You can click “Restart Service” on the home page.

📝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 info, success, warning, and error.
  • Keyword search: enter keywords to quickly locate specific logs.
  • Clear logs: remove all historical log records with one click.

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: A full deployment usually takes 3-10 minutes, depending on network speed and system performance.

👑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 and choose “Open membership”.
  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: Paid membership is tied to a device code, so you can purchase and use it without logging in. If you need cross-device management, you can log in and bind it to your account.

❓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 whether the port is occupied by another program (default 18789).
  2. Confirm OpenClaw is installed correctly and try redeploying if needed.
  3. Check the "Log Center" for detailed error messages.
  4. If the problem persists, contact customer service.
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 and choose “Redeploy” to re-enter the deployment wizard. Your existing configuration will be preserved so you can modify it if needed.

🎧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