Unified API for your workspace
Windows, sessions, layouts, and a local API for keeping the whole Mac organized and programmable.
Explore Lattices →An API for agents to use your computer.
Give agents a workspace of their own. Follow their work in a live preview.
macOS + Linux · CLI + MCP
A real terminal. A real browser. A live agent preview.
Open full-size video 29-second development preview. Scripted sequence, real MCP calls.
What happens in the demo
A Ghostty terminal sits on the left and a browser on the right of a virtual desktop. The terminal runs an Action call that opens the Lattices docs. Its next call creates a second virtual workspace, whose live preview appears over the browser. Further calls navigate that workspace to the docs and read the page title. The entire stage is recorded locally on a Mac. This is a timed script executing real calls, not an agent deciding its next steps live. There is no audio.
Try Action on your Mac
Install Action, grant permissions, and take your first screenshot.
macOS on Apple Silicon · Install Bun first
No source build needed.
Install and launch
Expect: Action.app opens from Applications. The CLI is available as lats.
bun add -g @arach/lattices lats action install --launchAllow access
Expect: Approve Accessibility and Screen Recording in macOS, then run the second command. Both permissions should report granted.
lats action call permissions.request lats action call permissions.snapshotTake a screenshot
Expect: The reply reports status: screenshot. Your main display capture opens in Preview. This captures your screen, not a private workspace.
lats action call capture.screenshotScreen \ '{"output":"/tmp/action-first.png"}' open /tmp/action-first.png
Something not working?
Agent not reachable? Run lats action launch, then retry. Permission denied? Enable Action in System Settings → Privacy & Security, quit and reopen Action, then check permissions again.
Next, connect your agent
The app download includes native capture. Agent-driven browser work and the virtual workspaces in the demo also need the matching MCP runtime and host preflight.
MCP setup and agent instructionsWorking with Linux?
Linux uses a remote lattices-host connection, not the macOS app. Available actions depend on the host; macOS overlays and Accessibility are not available there.
Remote Linux setupAppKit lifecycleScreenCaptureKitAX + OCRCLI + MCP
One local path fromintent to evidence.
An agent or operator calls the local Action runtime, which owns the session, targets, and orchestration. ActionAgent bridges those requests into native macOS work.
Action.app owns AppKit, WebKit, permissions, and capture. The run comes back with its receipts attached, including an explicit finished marker when recording is actually complete.
System · Four local planes, one inspectable run
One family, two focused products
Action and Lattices.
observeactverify
An API for computer use
Target resolution, on-device actions, capture, trace, and review for one inspectable piece of work.
Download Action →Use Action on its own, or pair it with Lattices when the whole workspace needs to stay organized and programmable.
Built for macOS
See what happened.
Review the video, screenshots, and activity log after a run. Action also records when capture has finished.
Find the right control
Action finds buttons and fields through Accessibility and the browser DOM before falling back to screen coordinates.
Run on the Mac
Action uses AppKit and ScreenCaptureKit, with access controlled by your macOS permissions.
Review the recording
Each run keeps its video, screenshots, activity log, and capture completion status together.
From source
Build it. Launch it. Prove it.
Local commands live in package.json. Run them with bun from products/action.
- macOS on Apple Silicon
- Bun and the Swift toolchain
- Accessibility permission
- Screen Recording permission
$ bun install$ bun run native:app:buildAction.app signed and ready$ bun run native:doctorAccessibility: granted · Screen Recording: granted$ bun run native:launchAction.app running