Skip to content

Latest commit

 

History

History
81 lines (66 loc) · 3.6 KB

File metadata and controls

81 lines (66 loc) · 3.6 KB
title ADE — Agentic Development Environment
description Orchestrate parallel AI coding agents from a unified desktop control plane. Lanes, missions, context packs, and conflict detection — all local-first.
New to ADE? Start here. Learn what it does, install the app, and set up your first project in 5 minutes. Explore every tab — Run, Lanes, Chat, Files, Graph, PRs, Missions, CTO, and more. AI providers, integrations, permissions, budgets, and project configuration. What's new in ADE — latest features, improvements, and fixes.

Why ADE?

AI coding agents are fast individually. Running them in parallel is where things break:

Two agents edit the same files and you only find out at merge time. Agents start cold instead of continuing informed from where they left off. Sessions, branches, costs, and artifacts live in separate tools with no unified view. You can launch work but cannot track quality, ownership, or state across agents. **ADE solves this** with isolated lanes, durable context packs, structured missions, a coordinator agent (CTO), and a workspace-wide view of branch state and merge risk.

Feature highlights

Isolated git worktrees, each with their own terminal sessions, agent context, and environment. Run agents in all of them simultaneously.
[Explore Lanes →](/lanes/overview)
Maintains project memory, delegates work to specialized workers, syncs with Linear, and manages budgets across all lanes.
[Explore CTO →](/cto/overview)
Multi-step execution with planning, approval, validation, cost tracking, and full audit trails.
[Explore Missions →](/missions/overview)
Trigger-driven rules that launch agent sessions or missions in response to git events, schedules, and webhooks.
[Explore Automations →](/automations/overview)
Infinite canvas showing every lane, branch, PR, and dependency. Run merge simulations visually and spot risks at a glance.
[Explore Graph →](/tools/workspace-graph)
Agents capture screenshots, short recordings, and ingested browser traces through the `ade proof` CLI. ADE normalizes and routes each artifact to its mission, chat session, PR, or Linear issue.
[Explore Proof →](/computer-use/overview)