Get started
Sign in with email and password, Google, or Apple. Connecting a CLI-based coding agent still requires a computer.1
Choose an environment
Pick the environment where the agent should work.
2
Choose an agent
Select a coding agent, model, and effort level. Goal mode appears when the selected agent supports it.
3
Add context
Type or dictate your prompt. When you dictate, the app sends the speech model a list of names from your code, so it transcribes them correctly. A recording can be up to 20 minutes long. You can attach photos, files, or a picture from the camera.
4
Start the task
Send the prompt. The workspace opens while it is being prepared, and Replicas sends your message when provisioning finishes.