Automations, agents & monitors

Agent guardrails that can never be softer than the client's policy

An agent is only as safe as its limits. Guardrails keep each agent within bounds you choose — and the wizard refuses to save guardrails weaker than the client's own policy.

What it does

Guardrails include the maximum budget change percentage, the maximum number of proposals per run, “reversible changes only”, whether the agent may move money (increase budgets), quiet hours after a human change, a daily cap and the minimum number of days of data before acting.

Next to the guardrails the wizard shows the client's current policy. If a guardrail would be softer — for example a higher budget change than the client allows — saving is blocked with a warning. Protected campaigns are never touched by any rule.

Why it helps

  • Agents cannot surprise you with large changes.
  • Consistency between agent settings and client policy.
  • Protect sensitive campaigns explicitly.

How to set it up

  1. In the agent wizard, open the guardrails step.
  2. Set limits, quiet hours, daily cap and minimum data days.
  3. Mark protected campaigns in the scope step.
  4. Save; fix any warning about guardrails softer than the policy.

Good to know

  • Under the auto policy, agents' reversible proposals within limits may apply automatically.
  • Every agent action is logged with the agent and run.
Ace, the AdAce Ads mascot

Try it on your own accounts

Create a workspace, connect Google or Meta in a couple of clicks and see your accounts clearly. Changes follow your approvals or the policy you configure.

Create your workspace