Quickstart

Get your first AI agent deployed in under 5 minutes.

Using Claude or another AI assistant? Copy a ready-made prompt that teaches it to install the CLI, authenticate, and deploy agents for you.

Installs the CLI from npm, lists the available templates, creates an openclaw agent named cli-demo, checks its status, lists all agents, and sends it a first chat message. Recorded against production.

1. Install the CLI

npm install -g maritime-cli

2. Authenticate

maritime login

Don't have an account yet? Sign up directly from the CLI:

maritime signup

3. Browse templates

See what agent templates are available:

maritime templates

# Output:
# NAME                FRAMEWORK            DESCRIPTION
# OpenClaw Agent      openclaw             Autonomous AI agent with 100+ skills.
# OpenClaw Browser    openclaw_browser     OpenClaw with Playwright MCP browser control.
# Hermes Agent        hermes               Self-improving Python agent with memory.
# ZeroClaw Agent      zeroclaw             Ultra-lightweight Rust agent.

The New agent dialog in the Maritime dashboard showing the template picker — The same templates power the dashboard's New agent flow. The CLI and the dashboard create identical agents.

4. Create an agent

Create a new agent from a template. You can pass options inline or follow the interactive prompts:

maritime create my-agent --template openclaw

# Or run without arguments and follow the prompts:
# maritime create

The same create flow in the dashboard: pick the OpenClaw template, name the agent, and watch the deploy pipeline run until the agent turns Active.

5. Set environment variables

Add any API keys your agent needs:

maritime env set my-agent OPENAI_API_KEY=sk-...

The Env tab of an agent showing a masked OPENAI_API_KEY variable — Variables are marked secret by default: stored encrypted, always masked in the UI. Uncheck Secret for plain-text config values.

6. Deploy

maritime deploy my-agent

An agent's Overview tab showing status, uptime, CPU, memory, and a live console — The agent's Overview tab after deploy: live status, uptime, CPU and memory, and a console preview.

7. Monitor

View your agent's logs:

maritime logs my-agent

The Console tab streaming live container logs for an agent — The same log stream in the dashboard's Console tab, alongside an interactive terminal and the agent's files.

Shipping a web app instead?

Any repo with a Dockerfile can be deployed as a public web agent: Maritime builds it, runs it serverlessly, and serves it on a public no-login URL. It sleeps when idle and wakes on the next visit:

maritime create my-app --repo https://github.com/you/app --public --port 3000

See the CLI Reference for all create options.

What happens next?

Your agent starts immediately as active. It counts against your plan's included agents from the moment it exists, and how much it runs never changes the bill. Use maritime sleep my-agent to park it when you don't need it running; sleeping agents auto-wake when triggered via webhook, cron schedule, or API call.