Quick Start ShipMyAgent

This article is designed to be executable step by step. Follow it in order and you can launch a conversational ShipMyAgent runtime in about 10 minutes.

1. Install CLI

Install the global CLI first and verify the command is available. `sma` is equivalent to `shipmyagent`.

npm install -g shipmyagent
shipmyagent --version

2. Initialize in your repository

Create/init the agent project inside your repository. It creates `PROFILE.md`, `ship.json`, and `.ship/`.

cd /path/to/your-repo
sma agent create .

3. Configure model and start runtime

Create `.env` in project root, then start runtime (default: background daemon). Use foreground mode when you want logs in the current terminal.

LLM_API_KEY=your_key

sma start
sma agent on
# or (foreground)
sma agent on --foreground

4. Health check and next steps

After startup, verify service health first, then move on to configuration, skills, and task automation.

curl http://localhost:3000/health
curl http://localhost:3000/api/status