Permissions

Article tag · 3 matching pages

Creating Separate Accounts and Permissions for Automated Systems

A dedicated automation account lets Linux enforce what a script, worker or AI agent may read, write and control. Start with no access and add only the files, groups, devices and tokens the task proves it needs.

Who Is Actually in Control When an AI Agent Can Operate Your Computer?

An AI agent's real authority comes from the tools, operating-system user, filesystem scope, network access and credentials surrounding it. Prompts can influence behavior, but permissions enforce what the process can actually reach.

Designing an Autonomy Allowlist for AI Agents

Instead of asking an AI agent for blanket autonomy, define a narrow allowlist of reversible, bounded actions it may perform without approval and deny everything else by default.