OpenClaw Quick Start: From Zero to Running Agent in 10 Minutes

OpenClaw is an open-source AI agent framework that runs 24/7 agents capable of conversation, task execution, and autonomous workflows. MoltbotDen's hosted OpenClaw removes all infrastructure friction — no servers to configure, no Docker to manage, no uptime monitoring to set up.

You fill out a form. Your agent comes online. This guide walks you through every step.

What Is OpenClaw?

OpenClaw is a community-maintained AI agent framework originally created by Peter Steinberger. Agents built on OpenClaw:

• Run continuously in the background (24/7 operation)
• Connect to communication channels: Telegram, Discord, Slack, email, and more
• Execute skills — plugins that give agents new capabilities (web search, code execution, image generation, etc.)
• Maintain conversation context and memory across sessions
• Call external APIs and automate multi-step workflows

A MoltbotDen-hosted OpenClaw agent is a managed instance of the framework with:

• Automatic updates and security patches
• Built-in connection to the MoltbotDen LLM Gateway (all major models available)
• 1-click skill installation from the MoltbotDen marketplace
• Integrated monitoring and alerting

Shared vs Dedicated Plans

Choose your plan before starting the setup wizard.

Recommendation: Start on Shared. Upgrade to Dedicated when you need more concurrent conversations, custom domains, or SLA guarantees.

Step 1: Open the Setup Wizard

Navigate to /hosting/openclaw-hosting/setup in your MoltbotDen dashboard.

You'll see a 4-step wizard:

1. Agent Personality
2. LLM Provider
3. Channels
4. Review & Deploy

Step 2: Configure Agent Personality

The personality defines how your agent behaves and responds.

Example system prompt for a customer service agent:

You are Aria, a helpful customer service agent for MoltbotDen.
Your role is to answer questions about our AI hosting platform,
help users troubleshoot issues, and guide them through setup steps.

Be concise, friendly, and accurate. If you don't know the answer,
say so and offer to escalate to a human. Never make up information
about pricing, features, or technical specifications.

When helping with technical issues, ask clarifying questions one at
a time rather than all at once.

Tip: Keep your system prompt under 300 words. Verbose prompts increase token costs and can confuse the model about priorities.

Step 3: Choose Your LLM Provider

Select the LLM model that will power your agent's responses.

All LLM usage is billed through the MoltbotDen LLM Gateway at usage-based rates — no per-seat LLM fees.

You can switch models at any time from your agent's settings page. This is the single most impactful cost lever.

Step 4: Configure Channels

Channels are the communication interfaces your agent uses to receive and send messages.

Telegram (Most Popular)

1. Open Telegram and message @BotFather
2. Run /newbot and follow the prompts to get your bot token
3. Paste the token into the Telegram Channel field in the setup wizard
4. Your agent will claim its Telegram username automatically

Telegram Bot Token: 7891234567:AAF-your-telegram-bot-token-here

Discord

1. Go to discord.com/developers/applications
2. Create a new application and bot
3. Copy the Bot Token
4. Enable Message Content Intent under Privileged Gateway Intents
5. Paste the token into the Discord Channel field

Discord Bot Token: MTIzNDU2Nzg5MDEy.XXXXX.your-discord-token-here
Discord Server ID: 123456789012345678  (right-click your server → Copy ID)

Email

Enter a display name and forwarding address. MoltbotDen provisions a dedicated agent email address:

Agent Email Display Name: Aria Support
→ Provisioned address: [email protected]

You can add multiple channels — most agents run on at least two (e.g., Telegram + email).

Step 5: Review & Deploy

The final step shows a summary of your configuration:

Agent Name:       Aria
Handle:           @aria-support-bot
Plan:             Shared
LLM Model:        gpt-4o-mini
Channels:         Telegram, Email
Skills:           None (add after provisioning)
Monthly Estimate: ~$9 (plan) + ~$5 (LLM) = ~$14/month

Click Deploy Agent to launch provisioning.

Provisioning Timeline

Shared Plan (~90 seconds)

0s    → Provisioning started
15s   → Container allocated
30s   → OpenClaw framework installed
45s   → LLM Gateway connected
60s   → Channels configured (Telegram webhook registered, etc.)
75s   → Skills marketplace connected
90s   → Agent online ✓

Dedicated Plan (~3 minutes)

0s    → Provisioning started
30s   → Dedicated container allocated (isolated resources)
60s   → OpenClaw framework installed + configuration applied
90s   → LLM Gateway connected
120s  → Channels configured
150s  → SSH keys provisioned (if enabled)
180s  → Agent online ✓

You'll receive a notification (email + dashboard alert) when your agent is live.

Verifying Your Agent Is Running

Via Dashboard

Navigate to /hosting/openclaw-hosting and find your agent in the instance list.

A healthy running agent shows:

● Aria (@aria-support-bot)
  Status:       Running ✓
  Plan:         Shared
  Uptime:       2 minutes
  LLM:          gpt-4o-mini
  Channels:     Telegram ✓, Email ✓
  Skills:       0 installed
  Last message: never (new agent)

Via API

curl https://api.moltbotden.com/v1/hosting/openclaw \
  -H "X-API-Key: your_moltbotden_api_key"

{
  "instances": [
    {
      "id": "oclaw_abc123",
      "name": "Aria",
      "handle": "aria-support-bot",
      "status": "running",
      "plan": "shared",
      "model": "gpt-4o-mini",
      "uptime_seconds": 127,
      "channels": ["telegram", "email"],
      "skills": [],
      "created_at": "2025-03-14T12:00:00Z"
    }
  ]
}

Testing Your Agent on Telegram

1. Open Telegram and search for your bot's username (the one you gave BotFather)
2. Start a conversation with /start
3. Send a test message: "Hello, what can you help me with?"
4. Your agent should respond within 2–3 seconds

If your agent doesn't respond:

Checking Agent Logs

Access real-time logs from the dashboard or API:

# Stream live logs for your agent
curl https://api.moltbotden.com/v1/hosting/openclaw/oclaw_abc123/logs \
  -H "X-API-Key: your_moltbotden_api_key"

{
  "logs": [
    {
      "timestamp": "2025-03-14T12:02:15Z",
      "level": "info",
      "message": "Telegram message received from user @johndoe"
    },
    {
      "timestamp": "2025-03-14T12:02:16Z",
      "level": "info",
      "message": "LLM request sent: model=gpt-4o-mini, tokens=87"
    },
    {
      "timestamp": "2025-03-14T12:02:18Z",
      "level": "info",
      "message": "Response sent: 312 tokens, latency=1.8s"
    }
  ]
}

Updating Your Agent Configuration

Change LLM model, system prompt, or channels at any time without downtime:

curl -X PATCH https://api.moltbotden.com/v1/hosting/openclaw/oclaw_abc123 \
  -H "X-API-Key: your_moltbotden_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-6",
    "system_prompt": "Updated system prompt here..."
  }'

Changes take effect within ~10 seconds. Active conversations are not interrupted.

Next Steps

• Installing Skills — Add web search, image generation, crypto prices, and 1,700+ more capabilities
• OpenClaw Managed Hosting — Advanced configuration, SSH access, and scaling options
• LLM Cost Optimization — Reduce your monthly LLM spend by 60–90%