Getting Started

Get up and running with DaoFlow in under 5 minutes. DaoFlow is an agent-first hosting platform for deterministic Docker systems — designed so AI coding agents and humans can deploy, inspect, and manage applications safely.

Prerequisites

Requirement	Minimum Version
Bun	1.3.9
Docker Engine	20.10+
Docker Compose	v2.0+
Git	any recent
Temporal CLI	recommended

Quick Start

# Clone the repository
git clone https://github.com/DaoFlow-dev/DaoFlow.git
cd DaoFlow

# Install dependencies
bun install

# Start local infrastructure
bun run dev:infra

# Push the schema into the local dev database
bun run db:push

# Start the API + web UI
bun run dev

# Optional: enable Temporal-backed workflows in a second terminal
bun run dev:temporal

Local endpoints:

• API server: http://localhost:3000
• Vite web UI: http://localhost:5173
• Temporal UI: http://localhost:8233 when bun run dev:temporal is running

What's Included

DaoFlow ships as a monorepo with four packages:

Package	Description
packages/server	API server (tRPC + Better Auth + Drizzle ORM)
packages/client	Web dashboard (React + Vite + shadcn/ui)
packages/cli	Agent-first CLI (daoflow command)
packages/shared	Shared types, scopes, and utilities

First Steps

1. Installation — Detailed environment setup
2. First Deployment — Deploy your first app
3. Configuration — Customize your instance
4. Staging Runbook — Rehearse production bring-up before going live

For AI Agents

If you're an AI coding agent, start with the CLI reference or the Agent Integration guide. The CLI supports --json output on every command for structured machine-readable responses.

# Install the CLI
curl -fsSL -o /usr/local/bin/daoflow \
  https://github.com/DaoFlow-dev/DaoFlow/releases/latest/download/daoflow-$(uname -s | tr A-Z a-z)-$(uname -m | sed 's/x86_64/x64/;s/aarch64/arm64/')
chmod +x /usr/local/bin/daoflow

# Authenticate
daoflow login --url http://localhost:3000 --token YOUR_TOKEN

# Check your permissions
daoflow capabilities --json