Introduction

Start is a desktop workspace for orchestrating AI coding agents. Instead of juggling a terminal, an editor, and a chat window, Start gives you a single environment where multiple Claude agents can work on your codebase simultaneously — while you stay in control.

Why orchestration matters

A single AI agent can write code, but real development involves coordinating across multiple concerns — refactoring a module while fixing tests, reviewing changes while planning the next feature. Start lets you run multiple agents in parallel, each with its own chat session, all operating on the same project.

What you can do with Start:

• Run parallel chat sessions, each with an independent Claude agent
• Manage work on a Kanban board that agents can read and update autonomously
• Let autopilot pick up tasks from the board and execute them without intervention
• Schedule recurring jobs — security scans, code reviews, doc audits — on a CRON schedule
• Persist project knowledge in shared memory so every agent starts with full context
• Monitor everything from your phone via Telegram integration

All data stays local on your machine. Start requires a Claude API key.

Get started