Next steps

Deploy your agent

Your config files are ready. Follow these steps to get your agent running locally, then deploy it anywhere.

1

Copy config files to your project

Create these files in the root of your project directory. If you used the wizard, you already have the contents copied to your clipboard.

# Create the files in your project root
mkdir -p my-agent && cd my-agent

# Paste the contents from the wizard into each file:
# - SOUL.md       (agent personality & boundaries)
# - AGENTS.md     (model, tools, permissions)
# - HEARTBEAT.md  (health checks & monitoring)
2

Install the OpenClaw CLI

The CLI validates your config files and starts your agent locally before deploying.

npm install -g openclaw

# Verify installation
openclaw --version
3

Start your agent

Run the start command from your project directory. OpenClaw reads your config files and boots the agent.

cd my-agent

# Start the agent in development mode
openclaw start

# You should see:
# [openclaw] Reading SOUL.md...
# [openclaw] Reading AGENTS.md...
# [openclaw] Reading HEARTBEAT.md...
# [openclaw] Agent "YourAgentName" is running on http://localhost:3100
4

Verify your agent is running

Check the health endpoint and test a basic interaction to make sure everything is wired up.

# Check health endpoint
curl http://localhost:3100/health

# Expected response:
# {"status":"ok","agent":"YourAgentName","uptime":"0m 12s"}

# Send a test message
curl -X POST http://localhost:3100/chat \
  -H "Content-Type: application/json" \
  -d '{"message": "Hello, are you running?"}'

Want one-click deploy instead?

Join the waitlist to get access to ClawSprout's one-click deploy to AWS, DigitalOcean, and Fly.io — no CLI required.

Join Waitlist