Sultan

Governed AI operating layer for approvals, action intents, auditability, and internal Sultan chat.

Command

OpenClaw

Approvals

This is the human gate. Agents can prepare and route work, but risky movement still stops here first.

Approval Packets

Chronicle-style human review queue.

Kernel

0

AI Commands Waiting

Suggestions that have not crossed into execution law.

Commands

0

Action Intents Waiting

Staged external moves still sitting in the human gate.

Intents

0

Approval Packets

Who needs to approve, why it matters, and what evidence exists.

0 waiting
Nothing is waiting right now. When risky work stacks up, this queue should become the clean human gate.

AI Commands

Suggestions that need approval before an action intent should move.

0 waiting

Action Intents

External work already staged and waiting on a human gate.

0 waiting