# AI use cases for Cua (trycua/cua) Repo: [trycua/cua](https://github.com/trycua/cua) | Use cases: 3 | Page: https://dedupedai.com/use-cases/trycua-cua/ Cua gives AI agents computers they can use. The repository holds Cua Spaces (a desktop app that gives agents full macOS or Linux desktops on your Mac, other machines you own or your own cloud), Cua Driver for operating native apps and browsers, Lume for local macOS and Linux VMs on Apple Silicon, a Cua SDK and CLI for sandboxes, CUA-S1 decision models and Cua Bench for evaluating computer-use agents. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Spin up a disposable Linux sandbox for an agent: https://dedupedai.com/use-cases/trycua-cua/spin-up-a-sandbox-for-an-agent/ - Let an agent operate native desktop apps without taking your mouse: https://dedupedai.com/use-cases/trycua-cua/let-an-agent-operate-desktop-apps/ - Create a local macOS VM on Apple Silicon with Lume: https://dedupedai.com/use-cases/trycua-cua/create-a-local-macos-vm-with-lume/ ## How to spin up a disposable Linux sandbox for an agent with Cua Repo: [trycua/cua](https://github.com/trycua/cua) | Difficulty: Beginner | Time: 15 minutes | Basis: From the README Install the Cua CLI and create, use, screenshot and remove a gVisor container in four commands. **Who it is for:** A developer who wants agents to run commands somewhere other than their own machine. **The problem:** Letting an agent run shell commands on your laptop risks your files and setup. **The outcome:** A disposable sandbox you can exec into, screenshot and delete, from the CLI or Python. ### Steps 1. Install with `curl -fsSL https://cua.ai/install.sh | sh` (Windows uses the PowerShell installer) 2. Create a sandbox: `cua sb create ubuntu --name dev` 3. Run a command inside: `cua sb exec dev uname -a` 4. Take a screenshot with `cua sb screenshot dev` 5. Remove it when done with `cua sb rm dev` ### Prompt to paste into your agent ```text Using trycua/cua, create a sandbox named dev, run inside it with cua sb exec, capture a screenshot, return the outputs to me, and then remove the sandbox. ``` ### Caveat Risk, remote script: The installer pipes a script from Cua.ai into your shell and runs Cua auth login; read it first if you prefer. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/trycua-cua/spin-up-a-sandbox-for-an-agent/ ## How to let an agent operate native desktop apps without taking your mouse with Cua Repo: [trycua/cua](https://github.com/trycua/cua) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Install Cua Driver so Claude Code, Codex or Cursor can inspect and operate apps in the background. **Who it is for:** A developer automating work in desktop apps that have no API. **The problem:** Most desktop automation hijacks your pointer and focus, so you cannot keep working. **The outcome:** An agent that operates apps and browsers through the CLI, MCP or SDKs, in the background where the platform supports it. ### Steps 1. On macOS or Linux run `/bin/bash -c "$(curl -fsSL https://cua.ai/driver/install.sh)"` 2. Follow the Drive your first app quickstart to grant permissions and connect your agent 3. Ask the agent to compute 6 x 7 in Calculator and verify that the app displays 42 4. Move on to a real task in your own app 5. Check platform support for which apps allow background delivery ### Prompt to paste into your agent ```text Install trycua/cua Cua Driver, connect it to , run the Calculator check (6 x 7 = 42), then use it to . Do not send, delete or purchase anything without asking me. ``` ### Caveat Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/trycua-cua/let-an-agent-operate-desktop-apps/ ## How to create a local macOS VM on Apple Silicon with Lume with Cua Repo: [trycua/cua](https://github.com/trycua/cua) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README Install Lume, create a macOS Tahoe VM from an Apple restore image and connect over SSH. **Who it is for:** A developer who needs clean macOS environments for testing or for agents to work in. **The problem:** Testing on a fresh macOS install or isolating an agent usually means a second Mac. **The outcome:** A vanilla macOS Tahoe VM running locally on Apple's Virtualization Framework, reachable over SSH. ### Steps 1. Install Lume: `/bin/bash -c "$(curl -fsSL https://cua.ai/lume/install.sh)"` 2. Follow the Create your first Lume VM quickstart 3. Create a macOS Tahoe VM from an Apple restore image and start it 4. Connect to it over SSH 5. Hand the VM to your agent or test suite instead of your main machine ### Prompt to paste into your agent ```text Using trycua/cua Lume on this Apple Silicon Mac, create a macOS Tahoe VM, start it, connect over SSH, and run inside it, then report the result. ``` ### Caveat Lume needs Apple Silicon, and a macOS VM image is a large download. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/trycua-cua/create-a-local-macos-vm-with-lume/