Quick Start
Get DevTeam Orchestrator running in 5 minutes. This guide walks you through installation, configuration, and running your first AI agent task.
Prerequisites
- Node.js 18 or later
- npm or pnpm
- An API key from your DevTeam instance (or a self-hosted setup)
If you do not have a DevTeam instance yet, see the Deployment Guide for self-hosting instructions.
1. Install the SDK and CLI
npm install devteam-sdk
npm install -g devteam-orchestrator-cli2. Configure Your Environment
Create a .env file in your project root:
.env
DEVTEAM_API_URL=https://devteam.marsala.dev
DEVTEAM_API_KEY=your-api-key-hereOr authenticate via the CLI:
devteam login
# Opens browser for JWT authentication
# Stores token in ~/.devteam/credentials.json3. Run Your First Task
hello.ts
import { DevTeamClient } from 'devteam-sdk';
const client = new DevTeamClient({
apiUrl: process.env.DEVTEAM_API_URL,
apiKey: process.env.DEVTEAM_API_KEY,
});
async function main() {
const task = await client.createTask({
prompt: 'Summarize the key benefits of microservice architecture in 3 bullet points.',
model: 'sonnet',
});
console.log('Task ID:', task.id);
console.log('Status:', task.status);
// Poll for completion
const result = await client.waitForTask(task.id, { timeoutMs: 30_000 });
console.log('Result:', result.output);
}
main().catch(console.error);Run it:
npx tsx hello.ts4. Check Task Status
Every task returns a status you can query:
const status = await client.getTaskStatus('dt_task_abc123');
console.log(status);
// {
// id: 'dt_task_abc123',
// status: 'completed', // pending | running | completed | failed | awaiting_approval
// model: 'sonnet',
// queue: 'default',
// createdAt: '2026-02-20T10:00:00Z',
// completedAt: '2026-02-20T10:00:03Z',
// output: '...',
// tokensUsed: { input: 42, output: 128 },
// }What's Next?
Learn the SDK
Explore the SDK Reference for the full TypeScript API including plans, fan-out, and HITL approvals.
Try Templates
Browse the Template Gallery and deploy a pre-built workflow in one command.
Set Up the CLI
Configure the CLI for power-user workflows and shell scripting.
Build a DAG Workflow
Follow the First Workflow tutorial to create a multi-step pipeline.