A process where risky agent actions are paused and routed to a human for review before execution.
Approval workflows are the mechanism behind human-in-the-loop. When the agent wants to take a risky action (send an email, process a refund, modify a database), the action is paused and a notification is sent to the designated approver.
The approver reviews the proposed action with full context — what the agent wants to do, why, and the potential impact. They can approve, reject, or modify the action. The decision is logged.
Clawctl blocks 70+ action categories by default. You can customize which actions require approval and which are auto-approved.
Approval workflows prevent the most dangerous AI mistakes. They provide a human checkpoint for high-risk actions while letting routine operations flow automatically.
Clawctl provides 70+ pre-configured approval categories. Approve via dashboard, Slack, or WhatsApp. Auto-approve rules for trusted patterns. 24-hour expiry by default. Full audit trail of all decisions.
Try Clawctl — 60 Second DeployApprovals expire after 24 hours by default. Most teams approve within minutes.
Yes. Approval notifications can be sent to Slack, WhatsApp, or any connected channel.
Yes. "Allow this forever" rules let you auto-approve trusted action patterns.
Human-in-the-Loop
A design pattern where an AI agent pauses before taking risky actions and waits for a human to approve or reject the action.
AI Guardrails
Safety boundaries that constrain what an AI agent can and cannot do, preventing harmful or unintended actions.
Audit Trail
A chronological record of every action an AI agent takes, providing accountability, compliance evidence, and forensic capability.
Agent Suspension
Temporarily disabling an AI agent so it stops processing messages and executing actions, without destroying its configuration or data.