OpenClaw Setup

Connect OpenClaw to your print farm so your personal AI assistant can monitor printers, control jobs, and manage hardware through natural conversation across WhatsApp, Telegram, Discord, Slack, and more.

Prerequisites

  • OpenClaw installed and running (curl -fsSL https://openclaw.ai/install.sh | bash)
  • A PrintHive API key (generate one in Settings → API Keys)

Configuration

Add the Hive MCP server to your ~/.openclaw/openclaw.json config file under the provider.mcpServers section:

{
  "provider": {
    "mcpServers": {
      "hive-mcp": {
        "type": "remote",
        "url": "https://api.printhiv3d.com/v1/mcp",
        "headers": {
          "x-api-key": "YOUR_API_KEY"
        }
      }
    }
  }
}

Replace YOUR_API_KEY with your actual API key.

Important: Use the headers field for authentication — do not put the API key as a query parameter in the URL. The MCP SDK's SSE transport drops query parameters on POST requests, which breaks authentication.

Verify Connection

After updating your config, restart OpenClaw and verify the server is connected:

openclaw mcp list

You should see hive-mcp in the list of connected servers.

Usage

Message your OpenClaw assistant through any connected channel (WhatsApp, Telegram, Discord, etc.) and ask about your printers:

> List all my printers and their status

Example Prompts

  • What printers do I have and which ones are currently printing?
  • Pause the print on my X1C
  • What filaments are loaded in the AMS?
  • Turn on the chamber light on Alphonse
  • Show me the temperature data for my last print job
  • Set the print speed to sport mode on all printers

Troubleshooting

  • Server not appearing in openclaw mcp list — Make sure the JSON config is valid (no trailing commas). Restart OpenClaw after editing the config file.
  • Authentication errors — Verify your API key is correct. Generate a new one in Settings → API Keys if needed.
  • Printer commands failing — Live printer control requires the printer to be online and connected via HiveLink. Check that HiveLink is running on your local network.