Skip to main content

​
What is ADE?

ADE (Agentic Development Environment) is a desktop application that serves as your DevOps cockpit for agentic coding workflows. While AI agents write code across multiple git branches simultaneously, ADE watches over them — tracking context, predicting conflicts, surfacing integration risks, and keeping everything synchronized. Think of ADE as the orchestration layer between you and your agents. You set the direction; ADE coordinates the execution.

Get Started in 5 Minutes

Download ADE, connect your repository, and launch your first parallel lane.

Understand Key Concepts

Learn about Lanes, Missions, the CTO agent, and how ADE thinks about parallel development.

Explore Features

Deep dives into Lanes, Missions, CTO, Computer Use, Automations, and more.

Tools & Views

Workspace Graph, File Editor, Terminals, PR management, Conflict detection, and History.

​
Core Philosophy

ADE is built around a single insight: agentic development is fundamentally parallel. Modern AI coding agents are fast, but they work best when they aren’t stepping on each other. ADE gives every agent its own isolated workspace (a Lane), keeps them all in sync, and surfaces conflicts before they become merge nightmares.

Parallel by Default

Every task gets its own git worktree. Agents never block each other.

Context-Aware Orchestration

The CTO agent understands your entire project and routes work intelligently.

Local-First Security

All data stays on your machine. API keys never leave the main process.

​
Feature Overview

Lanes are isolated git worktrees, each with their own terminal sessions, agent context, and execution environment. Create dozens of lanes and run agents in all of them simultaneously. Learn more
The CTO agent lives at the top of an org chart of worker agents. It understands your architecture, syncs with Linear, manages budgets, and delegates work to specialized employee agents. Learn more
Missions are the formal execution unit for complex multi-step AI work. A mission goes through Planning, Approval, Execution, and Validation, with full audit trails and cost tracking. Learn more
Trigger-driven rules with time-based and action-based execution that launch an automation session, mission, or built-in task. Learn more
ADE acts as the control plane for computer-use workflows. Connect Ghost OS or agent-browser backends, track proof artifacts, and route visual verification to missions, PRs, and Linear. Learn more
An infinite canvas showing every lane, branch, PR, and dependency relationship in your project. Run merge simulations visually and spot integration risks at a glance. Learn more

​
Configure ADE

Project Config

Set up your .ade/ directory, config layering, and trust model.

AI Providers

Configure Anthropic, OpenAI, OpenRouter, and local models.

Permissions

Trust boundaries, per-agent permissions, and audit trails.