Approvals & the safety gate
Turn on Nebula's safety gate to pause agents before risky actions — then approve, deny, or always-allow each write and delete from the prompt bar.
The safety gate pauses an agent before it runs an action that looks risky — an irreversible delete, a message to someone outside your team, or anything that doesn't match what you asked — and asks you to approve it first. It's off by default, so agents act without interruption until you switch it on.
Turning the safety gate on
The gate is a single on/off switch — there's nothing to configure beyond turning it on. You can flip it from two places.
From the prompt bar — click the shield icon to the right of the message input. Its tooltip reads Safety gate: on (pauses risky actions) or Safety gate: off (runs everything).
From settings — open Account → Profile and find the Write approval section. Toggle Safety gate on. The same switch drives both surfaces.
Off by default. When on, the gate applies to your actions in the current workspace — each person sets their own.
What the gate catches
When it's on, Nebula reviews each write or delete an agent is about to run and pauses the ones that look dangerous or unexpected: irreversible deletes, sending to people outside your team, or anything that doesn't match what you asked for. Everything else runs without a prompt.
Approving an action
When the gate fires, the agent stops and shows an approval card. The card names the action, explains why it was flagged, and lays out exactly what the action will do, so you can decide with full context.
| Approve | Deny | Always allow | |
|---|---|---|---|
| What it does | Runs this one action | Blocks it; the agent moves on | Runs it and stops asking for this tool |
| Add a note? | No | Optional reason sent to the agent | No |
| Best when | The action is right | It's wrong or you've changed your mind | You trust this tool and don't want to be re-asked |
If several actions are waiting, the card steps through them one at a time — "This agent wants to run a write/delete action (1 of N pending)."
Cards don't wait forever
Every card that asks you for something carries a five-minute countdown, ticking down on the card itself. Let it run out and the card flips to Expired.
An expired card behaves like a denial: the action doesn't run, the agent is told the request expired without a response, and it moves on with whatever it can still do. Answering after the countdown ends won't land — the decision is already made.
The countdown applies to every card that blocks an agent mid-run, not just the safety gate.
The safety gate card above — approve, deny, or always-allow a write or delete.
An agent needs a key or token it doesn't have yet.
An agent needs you to connect an app before it can act in it.
The agent needs you to pick between options before it continues.
If a card expires on you, just ask again — the agent picks the work back up, and this time the card is fresh.
Always-allowed tools
Choosing Always allow adds that tool to a per-workspace skip list, so future actions from it run without a prompt. Review and revoke the list anytime under Always allowed (skip the safety check) in the Write approval section — removing a tool puts it back under the gate.
Prefer the gate's reason over blanket trust. Approve case by case while you're learning what an agent does, then reach for Always allow only for the tools you've seen behave.
Related
Your profile
Set your Nebula display name, avatar, timezone, and appearance. Your profile follows you across every workspace you belong to — personal or team.
Connected accounts
Connect your own accounts — Gmail, Calendar, GitHub, and more — so your agents can act on your behalf. Connections stay private to you, per workspace.