What is an agentic development environment (ADE)?

An agentic development environment (ADE) is an app built around coding agents rather than around an editor. You give agents tasks; they read and change your code and run commands; and the ADE is where you start them, follow their work, answer their questions and review what they changed. Moontors is an ADE for Claude Code, Codex and Gemini CLI: it runs each vendor’s own agent, several at once, on macOS, Windows and Linux.

What an ADE does

An agent working on its own still needs you: to give it the task, to allow what it asks to run, and to judge the result. An ADE is built for that part of the work.

Several agents at once: each task gets its own session, and you see every session’s state without switching terminals.
Changes kept apart: agents working in one repository each get their own branch and folder, so one agent’s edits never land in another’s.
You stay in charge: the agent asks before it runs a command or edits a file, as much as you choose, and you answer where you’re looking.
Review before you keep: each session’s working changes are shown hunk by hunk, and any commit can be compared with them.
Work that doesn’t need the window: agents carry on while you do something else, or with the app closed.

Moontors as an ADE

Every session is a card on one board, streaming its work as it happens. See Sessions and the board.
A session can run in its own git worktree and branch. See Worktrees.
Permission modes, from asking before every edit to asking nothing, set per session. See Permission modes.
Each session’s repository in the Git view, and its files to read and edit. See Git and Files.
A background service runs the agents, so they keep working when the window is closed.
On your phone, an Android app on Google Play that follows and answers your computer's own sessions, end to end, with no Moontors server between them; the iPhone's coming soon. See The phone.
Each agent keeps its own sign-in, settings, MCP servers, skills and instruction files. See Configuration.
No Moontors account and no Moontors servers: each agent talks to its own vendor, as it would in your terminal. Moontors itself sends no telemetry; each agent's own usage data follows that agent's settings.

How it differs

From an IDE plugin

IDEs like VS Code and Cursor were built around your files, for writing code by hand. A plugin puts an agent in a panel beside the editor: the editor stays at the center, and the agent works in the folder the editor has open. An ADE puts the agents at the center: many sessions, each in its own folder or worktree, whatever editor you use.

From one agent’s own app

A vendor’s desktop app runs that vendor’s agent. Moontors runs Claude Code, Codex and Gemini CLI side by side, each signed in with its own account, so you can use each where it suits and compare their work on the same project.

From the agents in a terminal

The agents are the same: Moontors runs each vendor’s own official CLI. What it adds is everything around them: the board, worktrees, review, notifications and a background service, instead of a terminal window per agent.

Moontors doesn’t replace the agents or their accounts: you still need a plan or an API key with each vendor. See Signing in.

Questions

Is an ADE the same as an IDE?

No. An IDE is built around editing code, with an agent in one of its panels. An ADE is built around the agents: many sessions, each in its own folder or worktree, beside whatever editor you use.

Do I need to stop using my editor?

No. The agents work in folders on your computer, so your editor sees their changes as they make them, and Moontors shows each session’s changes in its Git view for you to review.

Which agents does Moontors run?

Claude Code, Codex and Gemini CLI, each as its vendor ships it and signed in with your own account. See Signing in.

Is Moontors free?

Yes, with no Moontors account. You use your own plan or API key with each agent’s vendor.