Native Mac app that runs a team of Claude Code and Codex agents in separate git worktrees, with review checks and a built-in browser for testing.
ManyAgents window — orchestrator, workers in git worktrees, built-in browser preview
Overview
What is ManyAgents?
Run a team of AI coding agents from one Mac window
ManyAgents is a native Mac app for running a team of Claude Code and Codex agents from one window. Each project gets an orchestrator that splits your request into tasks; workers build each task in its own git worktree, thinkers take the hard calls, and read-only reviewers check finished work before it reaches you. Agents test their own work in a built-in browser beside the chat, and can use an iOS simulator panel for mobile screens. It needs macOS 14 or later and Claude Code with a Claude subscription or Anthropic API key; Codex is optional. The trial runs 14 days with no account and no card; after that the app stays readable but new work pauses until you add a $49 one-time license.
Platforms and languages
ManyAgents Availability
Platforms
Languages
Capabilities
ManyAgents Key Features
One orchestrator per project
Tell the orchestrator what you want and it splits the request into tasks, opens a tab for each, and keeps track of who is doing what.
Workers in their own git worktrees
Each worker builds one task in its own worktree, tests it and reports back, so agents never trip over each other.
Read-only reviewers on another provider
Reviewers check finished work before it reaches you, and run on a different provider when Codex is set up, catching what self-review misses.
Built-in browser testing
Agents open your app in a browser beside the chat, click through flows, fill in forms and check the result themselves, including an iOS simulator panel.
Every question in one place
When an agent needs a decision it lands in one queue with its recommendation, so nothing waits unseen in a tab.
Uses the plans you already have
Agents run through your own Claude Code and Codex accounts, so there is no extra AI bill, and your code stays on your Mac with no analytics.
Best for
Who uses ManyAgents?
developers running several AI coding agents at once
coordinate Claude Code and Codex agents across projects without losing track of who is doing what
developers who want agents to test their own work
let agents click through the app in a built-in browser and report verified results
Plans and access
ManyAgents Pricing
Paid
Free for 14 days, no account or card; then a $49 one-time license (read-only until you add one).
Common questions
ManyAgents FAQs
What do I need to run ManyAgents?
A Mac on macOS 14 or later, and Claude Code with a Claude subscription or an Anthropic API key. Codex is optional.
What happens after the 14-day trial?
You can still read everything, but new work pauses until you add a license, which costs $49 paid once.
Is my code sent to ManyAgents?
No. Agents run through your own Claude Code and Codex accounts, and the app has no analytics.
How do agents test their work?
Your app runs in a browser beside the chat, where agents open it, click through flows, fill in forms and check the result themselves.
Reviews
See what the community thinks and share your experience.
—
Based on 0 ratings
Rating distribution
Community reviews
No reviews
No written reviews yet.
