What Is a Cloud Agent? Definition & Examples

A cloud agent is an AI agent that runs on a remote machine instead of your own computer. You give it a task through a web app, a chat message or an API call. It works in its own environment and reports back when it has a result.

The agent keeps running after you close your laptop. You can start several at once and check on them later.

What makes something a cloud agent

• It runs remotely. The model calls, the tools and the files all live on a server you don't manage.
• It works asynchronously. You hand off a task and come back to a result. You don't watch each step.
• It has its own environment. Coding agents usually get a fresh sandbox or VM per task, with the repo cloned and dependencies installed.

Cloud coding agents

Many cloud agents write code. They take a task, change the repository, run the tests and open a pull request.

• Hoplite runs each task in an isolated development environment. The agent edits the repo, runs tests, drives a browser when the app has a preview and opens a GitHub pull request. You can keep replying in the same thread as review comments come in.
• Capy lets you hand tasks to parallel coding agents that plan the change, build it and open a pull request.

General-purpose cloud agents

Not every cloud agent writes code. Grok Bot from xAI is a general-purpose example. Each bot gets its own cloud computer and signs into the tools you already use, like email and other apps. It works on multi-step tasks continuously and asks you only when it needs approval.

Cloud agent vs. local agent

A local agent, like a coding assistant in your terminal, runs on your machine. It uses your files and your credentials, and it stops when you close the terminal.

A cloud agent moves all of that to a server. You get less direct control, but you can run many tasks in parallel and your own machine stays free.