Skip to content
BENIBIX Learn

Agent Sandbox vs Agent Permissions

People often think that if an agent has strict permissions, it does not need a sandbox. Permissions decide which actions the agent may take, and they depend on each action being checked. A sandbox limits what any action can reach, even one that slipped through. Permissions stop the wrong step. A sandbox keeps the damage small when one gets past.

AI🤖 AI Agents

Agent Sandbox

A closed box where an agent can run code. It can't reach the real computer.

An agent sandbox is a closed space, like a container or a spare virtual machine, where an agent can run code without reaching the real system. Its walls set what the agent can touch: which folders, which websites and which secrets. If the agent makes a mistake or is tricked, the harm stays inside. So the agent can work with fewer stops to ask, since the worst case is small. Sandboxes and permissions work best as a pair.

Example
Jack lets his agent install things and run any command. But it all runs in a box with a copy of the project. His keys are not in the box.
Why it matters
Permissions only work if the agent asks. A sandbox keeps the harm small, even when a bad step gets through.
Full entry
AI🤖 AI Agents

Agent Permissions

Rules for an agent. They say what it can do alone, what it must ask about, and what it can't do.

Agent permissions are rules that say what an agent may do on its own. They also say what needs a person's yes, and what is never allowed. Rules are often set per tool or per command. For example, it may read files, must ask before it edits them and may never touch secret keys. The harness checks each action against the rules before it runs. Tight rules lower the risk, and loose ones mean fewer stops.

Example
Emma lets her coding agent read any file and run the tests on its own. It must ask before it edits code, and it can never read the passwords file.
Why it matters
An agent acts fast. Permissions decide which of its mistakes are possible at all.
Full entry

In one line

Agent Sandbox
A closed box where an agent can run code. It can't reach the real computer.
Agent Permissions
Rules for an agent. They say what it can do alone, what it must ask about, and what it can't do.