You just signed up for OpenClaw Cloud. Now what? This guide walks you through everything—from the welcome email to having a working AI assistant you can talk to from your phone. By the end, you'll have explored the dashboard, had your first conversation, tweaked the personality, and connected Telegram.

No terminal. No code. No API keys to hunt down. Just point, click, and chat.

Here's the Plan

0–1 min

Open welcome email → get your dashboard URL and password

1–2 min

Get Google API key → visit aistudio.google.com/apikey (free, 30 seconds)

2–3 min

Log in & connect → enter password and API key, hit Connect

3–5 min

First chat → send messages, see the AI respond in real time

5–8 min

Explore Config → customize name, personality, and behavior

8–10 min

Connect Telegram → chat with your AI from your phone

Let's go.

Minute 0–1: Your Welcome Email

After you purchase OpenClaw Cloud, you'll receive a welcome email within a few minutes. It contains two important things:

Your dashboard URL — something like your-name.clawdbot.com
Your temporary password — you'll use this to connect for the first time

📧 Screenshot: Welcome email showing dashboard URL and temporary password

Didn't get the email? Check your spam folder. If it's not there, give it 5 minutes—provisioning can occasionally take a moment. Still nothing? Reply to your purchase confirmation.

Minute 1–2: Get Your Google API Key

OpenClaw Cloud uses Google's free Gemini model. Before you can start chatting, you'll need your own Google AI Studio API key. Don't worry — it's completely free and takes 30 seconds.

1 Visit Google AI Studio

Go to aistudio.google.com/apikey and sign in with any Google account.

2 Create a new API key

Click Create API Key, then Create API key in new project. Copy the key that appears — you'll paste it into your OpenClaw dashboard in the next step.

Why Google? Google's Gemini model is powerful, free, and has generous rate limits. You're not locked in — you can always add Claude, GPT-4, or other models later via your own API keys.

Minute 2–3: Log In & Connect

1 Open your dashboard

Click the dashboard URL from your email (or type it into your browser). You'll land on the Overview tab.

🖥️ Screenshot: Overview tab showing the password field and Connect button

2 Enter your password and API key, then click Connect

Paste the temporary password from your welcome email and your Google API key from the previous step. Click Connect. The dashboard will connect to your running OpenClaw instance and you'll see the status turn green.

Your instance is already running. Unlike self-hosted OpenClaw, you don't need to install or start anything. The server is provisioned and running before you even log in. The password just connects your dashboard to it.

Minute 3–5: Your First Chat

You're connected. Time to talk to your assistant.

3 Open the webchat

Look for the chat widget on the right side of the dashboard (or the Chat tab, depending on your screen size). Type a message and hit send.

Try these first messages

"Hey! What can you do?"

"Tell me a joke"

"What's the weather like in New York?"

"Summarize the latest news about AI"

Your assistant will respond in real time. It streams the response word by word, just like ChatGPT.

🖥️ Screenshot: Webchat showing a conversation with the AI assistant

About Google Gemini: Your OpenClaw Cloud instance uses Google's powerful Gemini model with your free API key. No usage limits beyond Google's generous free tier. If you want to use different models (like Claude or GPT-4), you can add additional API keys later in Config.

Minute 5–8: Explore the Config Tab

The Config tab is where you make the assistant yours. Click Config in the dashboard navigation.

Things you can customize right now:

🏷️ Assistant Name

Give your assistant a name. "Jarvis," "Friday," "Claude," whatever you want. This is how it'll refer to itself.

🎭 Personality / System Prompt

This is the big one. The system prompt tells the AI who it is and how to behave. Want a professional assistant? A sarcastic friend? A writing coach? This is where you define that. Keep it concise—a few sentences works better than a novel.

Example system prompt

"You are Atlas, a sharp and efficient personal assistant. You're direct, occasionally witty, and always helpful. You keep responses concise unless asked to elaborate."

🧠 Model Selection

Google Gemini is already selected and works great for most tasks. If you add additional API keys (Anthropic, OpenAI, etc.), you can switch to Claude, GPT-4, or other models here.

🔧 Other Settings

Browse around—you'll find options for memory, context length, and more. Don't worry about getting everything perfect now. You can always come back and adjust.

🖥️ Screenshot: Config tab showing personality/system prompt field and model selection

After making changes, click Save. Your assistant will immediately start using the new settings.

Minute 8–10: Connect Telegram

The webchat is great, but the real magic is having your AI assistant in your pocket. Let's connect Telegram so you can chat from your phone.

4 Create a Telegram bot (1 minute)

Open Telegram, search for @BotFather, and send /newbot. Give it a name and username. BotFather will respond with a bot token—copy it.

5 Paste the token (30 seconds)

Back in your OpenClaw Cloud dashboard, click the Channels tab. Paste your Telegram bot token in the field and click Save.

6 Start chatting from your phone

Open your new bot in Telegram and send a message. Your AI assistant responds right there in Telegram. Done.

📱 Screenshot: Telegram conversation with the AI assistant bot

Want the full Telegram walkthrough? Check out our detailed guide: How to Connect Telegram to OpenClaw Cloud

What's Next?

You've got a working AI assistant with a custom personality, accessible from both the web dashboard and Telegram. Here are some ideas for what to explore next:

Connect Discord — Add your assistant to a Discord server. Full Discord setup guide →
Add your own API key — Upgrade to Claude, GPT-4, or other premium models via the Config tab
Explore skills — OpenClaw comes with skills for web search, file handling, code execution, and more
Set up memory — Let your assistant remember things across conversations
Build workflows — Create automated routines that run on a schedule

Quick Reference: Dashboard Tabs

Overview — Instance status, connection info, quick stats
Config — Personality, model selection, behavior settings
Channels — Connect Telegram, Discord, and other chat platforms
Chat — Built-in webchat to talk to your assistant directly
Files — Manage files your assistant can access
Logs — See what your assistant has been up to

FAQ

Do I need an API key to start?

Yes, but it's free. OpenClaw Cloud uses Google's Gemini model with your free Google AI Studio API key. Get one at aistudio.google.com/apikey in 30 seconds — no credit card required. Add additional keys later if you want premium models like Claude or GPT-4.

Can I use multiple channels at once?

Yes. Connect Telegram, Discord, webchat, and more—all at the same time. Your assistant's memory and personality stay consistent across all channels.

What if I mess up the config?

Nothing is permanent. Just go back to the Config tab and change it. Your assistant picks up the new settings immediately.

Is there a mobile app?

There's no dedicated app, but that's the beauty of connecting Telegram or Discord—those are your mobile apps. Your AI assistant lives wherever your chat apps are.

Ready to get started?

Your AI assistant is 60 seconds away. Free AI model — just add your Google API key.

Get OpenClaw Cloud — $24.99/mo

Cancel anytime · Free Google API key required · Learn more about OpenClaw Cloud