Computer Use

Overview

Computer use gives an agent a managed place to carry out interactive tasks.

Use it when the agent needs to do work that looks more like operating a computer than calling one simple tool.

The mental model is:

• a normal tool call is one clean action
• computer use is a small working environment where the agent can carry out a sequence of visible steps

That makes computer use powerful, but it also means it should be used more carefully than ordinary tools.

---

A concrete example

Imagine an agent needs to walk through a browser-based admin interface that does not have one clean API.

Computer use can help the agent:

1. open the interface
2. navigate through the relevant screens
3. gather or enter the needed information
4. report the result back into the thread or workflow

That is very different from a simple "call this endpoint" tool. It is closer to giving the agent a controlled workstation for a narrow task.

From the CLI, the operational loop looks like this:

archastro list agentcomputers --agent <agent_id>
archastro create agentcomputer --agent <agent_id> -n "ops-workstation"
archastro describe agentcomputer <computer_id>
archastro refresh agentcomputer <computer_id>
archastro exec agentcomputer <computer_id> -c "pwd"

That gives you a real workflow for provisioning, checking readiness, and validating the environment before you ask the agent to rely on it.

---

What computer use is not

Computer use is not the default way agents should operate.

It is the wrong choice when:

• one explicit API or tool call would do
• the task can be expressed as a clean workflow step
• the task is sensitive enough that a human should do it directly

The point is not to make agents click around for the sake of it. The point is to give them a controlled way to handle the cases where interactive work is genuinely necessary.

---

When it helps

Computer use is a good fit when an agent needs to:

• work through a multi-step interface
• inspect or manipulate a system that is not exposed as one clean API
• carry out a guided operational task

It is not the right first choice for everything. If a smaller, clearer tool will do, use the smaller tool.

The rule of thumb is simple: if you can express the action as one explicit tool, do that first. Reach for computer use when the work is genuinely interactive.

Built-in tools

Agents can use computers during conversations through four built-in tools:

computer_exec

Execute a shell command on the agent's computer.

Parameter	Type	Required	Description
command	string	Yes	Shell command to execute
working_directory	string	No	Working directory

Returns stdout, exit_code, and status.

computer_write_file

Write content to a file on the computer.

Parameter	Type	Required	Description
path	string	Yes	Absolute file path
content	string	Yes	File content to write

computer_read_file

Read the contents of a file from the computer.

Parameter	Type	Required	Description
path	string	Yes	Absolute file path

use_claude

Start an asynchronous Claude Code run on the agent's computer. This tool creates a durable agent session and returns immediately with a session ID instead of blocking the current turn.

Parameter	Type	Required	Description
prompt	string	Yes	Prompt to send to Claude
working_directory	string	No	Working directory for the run
name	string	No	Optional label for the durable Claude session

use_claude only resolves when the agent has computer use, an active GitHub installation, and sub-agents enabled. At execution time the session injects GitHub and Claude credentials when available, runs Claude non-interactively on the VM, and reports the result back through the existing durable sub-agent/session flow.

Tool resolution

When an agent uses computer tools, the platform automatically routes the request to a ready computer associated with that agent.

---

Safety guidelines

Computer use increases agent capability, so it needs stronger guardrails.

Before enabling computer use, be clear on:

• what environment the agent can use
• what actions are allowed
• what approvals exist
• how a human can review the result

From the CLI:

• inspect the computer status before using it
• keep the environment narrow
• destroy computers you no longer need

---

Best practices

1. Start with narrow tasks.
2. Keep the environment limited to the work at hand.
3. Put sensitive actions behind explicit approval.
4. Review outputs and logs regularly.
5. Prefer simpler tools when they are sufficient.

If you cannot explain why computer use is necessary for the task, it is not.