How we decide which agent actions need a human gate (a 3-question rubric we encode in the workflow)

"Add a human-in-the-loop" is easy to say and vague to implement. Which actions get a gate? We encode the decision as three checks per agent action, evaluated before the action runs. 1. Reversibility. Tag each action with a blast radius. Cheap-to-undo → can auto-execute. Irreversible or compliance-bearing (a submitted code, a denial) → force the gate. This is metadata on the tool, not a vibe. 2. Groundedness. The action must carry a citation from retrieval (policy clause, guideline). No source →…