Deploy CoPaw — Your Personal AI Assistant — to Your Own Server

CoPaw is your personal AI assistant — easy to install, deploy on your own machine or in the cloud. It connects to DingTalk, Feishu, QQ, Discord, iMessage, and more. With extensible skills, scheduled reminders, and support for both cloud and local LLMs, CoPaw is a warm companion always ready to lend a paw.

Setting it up manually involves installing Python, configuring providers, setting up channels, and managing the process. With HowToDeploy, you get a fully running CoPaw instance with its web console in minutes.

Why CoPaw?

• Every channel — DingTalk, Feishu, QQ, Discord, iMessage, and more from a single assistant
• Under your control — memory and personalization stay on your server, not someone else's cloud
• Extensible skills — built-in cron, custom skills auto-loaded from your workspace
• Local model support — run LLMs entirely on your hardware with Ollama, llama.cpp, or MLX
• Web console — chat, configure, and manage your assistant from a browser-based UI
• Scheduled reminders — set up heartbeat check-ins and digest notifications to any channel
• No lock-in — open-source, Apache 2.0 licensed

Prerequisites

Before you start, you'll need:

• A HowToDeploy account (sign up free)
• A cloud provider API key (DigitalOcean, Hetzner, Vultr, Linode, or AWS)
• An LLM provider API key (DashScope, OpenAI, or use local models)

Step 1: Connect your cloud provider

Go to Settings → Cloud Providers and paste your API key.

Tip: CoPaw runs well on just 2GB RAM when using cloud LLMs. If you plan to run local models on the server, choose a larger instance with GPU support.

Step 2: Deploy CoPaw

Head to the Dashboard and find CoPaw in the AI Agents section. Click the card to open the deploy form.

You need one field:

• DashScope API Key — the default cloud LLM provider for CoPaw (leave blank if using local models)

Server size, region, and Python setup are all pre-configured.

Step 3: Configure additional providers (optional)

Expand Advanced Settings to add alternative API keys:

• OpenAI API Key — use GPT models instead of DashScope
• Tavily API Key — enable web search capabilities for your assistant

Step 4: Add a custom domain (optional)

Want your assistant at ai.yourdomain.com? Enter your domain in Advanced Settings.

After deployment, point an A record for your domain to the server IP, click Verify DNS, and Caddy issues the SSL certificate automatically.

Step 5: Start chatting

Once deployment completes, open the CoPaw console in your browser. From there you can:

1. Chat directly — test your assistant in the built-in web UI
2. Connect channels — add DingTalk, Feishu, QQ, Discord, or iMessage from the Settings page
3. Add skills — drop Python scripts into the skills directory and they're auto-loaded
4. Set up heartbeats — schedule daily digests or periodic check-ins to any connected channel

Full channel setup guides are in the CoPaw documentation.

Supported channels

What's included

Every CoPaw deployment includes:

• Python runtime — pre-installed with all CoPaw dependencies
• Web console — browser-based UI for chat, settings, and skill management
• Persistent storage — memory, config, and skills survive restarts
• Caddy — automatic HTTPS with Let's Encrypt (when using a custom domain)
• Full SSH access — your server, your assistant, your data

Pricing

You pay your cloud provider directly for the server (typically $6-12/month for the recommended spec). HowToDeploy charges a small monthly management fee for monitoring and support.

Start with a 7-day free trial — no credit card required.

---

Ready to deploy your personal AI assistant? Deploy CoPaw now →