> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cloudthinker.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Approval Mode

> Choose Manual or Auto for the workspace, each run type, one Automation, or one chat, and see what each choice does

The Approval mode decides each write that no [Rule](/guide/approvals/rules) covers. Each run uses one mode: **Manual** or **Auto**.

## Why Approval mode

* **Rules first, mode second.** A Rule always decides first, in every mode.
* **One choice covers the rest.** Set the workspace mode once, and every run uses it.
* **Tighter where it matters.** A run type, an Automation, or one chat can pick its own mode.
* **No person, no wait.** Runs that no person can answer refuse instead of waiting.

## The modes

| Mode | What happens to a write |
| - | - |
| **Manual** | You approve every write, also one that an Allow Rule covers. Ask first and Deny Rules still apply. |
| **Auto** | Each write goes through three steps, in this order |

Under **Auto**, a write goes through these steps:

1. **Your Rules** allow, ask, or deny the write. This step is always on.
2. **AI judgment** allows a safe write that no Rule covers. You can turn this step off.
3. **If still not allowed**, the write either waits for a person (**Ask me**) or is refused (**Refuse**).

The picker names each Auto choice by its steps:

| Choice | AI judgment | If still not allowed |
| - | - | - |
| **Auto · Rules · Ask me** | Off | Asks a person |
| **Auto · Rules · Refuse** | Off | Refused |
| **Auto · Rules + AI · Ask me** | On | Asks a person |
| **Auto · Rules + AI · Refuse** | On | Refused |

Reads always run, in every mode. When AI judgment cannot finish its check, the write goes to step 3.

New workspaces start in **Auto · Rules + AI · Ask me**.

## Set the workspace mode

<Steps>
  <Step title="Open the Approval mode tab">
    Open **Approvals → Approval mode**.
  </Step>

  <Step title="Pick the mode">
    Under **Workspace mode**, open the picker. Select **Manual**, or select **Auto** and set **AI judgment** and **If still not allowed**.

    **Success state:** the picker shows your choice, such as **Auto · Rules + AI · Ask me**. It applies at once, with no confirm step.
  </Step>
</Steps>

You need the workspace Admin role to change a mode. Other roles see "View only. Ask a workspace admin to change these."

## Set a mode for a run type

A run type can use its own mode, also a looser one. Each row reads **Same as workspace** until you click **Override**. Click **Reset** to follow the workspace again.

| Run type | Default | Can ask a person |
| - | - | - |
| **Chat** | Same as workspace | Yes |
| **Resolve** | Same as workspace | Yes |
| **Optimize** | Same as workspace | No |
| **Automations** | **Auto · Rules + AI · Refuse** (built in) | Yes |
| **Review** | Same as workspace | No |

No person can answer an Optimize or Review run, so **Ask me** becomes **Refuse** there.

## Set a mode for one Automation or one chat

* **One Automation:** open the Automation and use its approval picker. **Default** follows the **Automations** run type.
* **One chat:** open the approval control in the chat prompt box, or type `/` and pick **Approval**. **Reset to default** follows the run type and workspace again.

A chat pick never changes the workspace mode. Before the first message, your pick is a draft, and the first message saves it. The agent is told about each change.

Under **Chat composer**, the switch **Let people pick AI judgment for one chat** limits what people can pick for one chat. When it is off, people can pick only **Manual** or **Auto** with AI judgment off. The workspace and run type modes still apply.

## Which mode a run uses

The most specific mode that someone set wins, even when it is looser:

1. The mode picked for this chat
2. The mode of the Automation that started the run
3. The mode of the run type
4. The workspace mode

For example, a workspace in **Manual** with **Automations** on **Auto · Rules + AI · Refuse** runs a scheduled Automation in Auto.

## FAQ

<AccordionGroup>
  <Accordion title="Does Manual turn my Rules off?">
    No. Ask first and Deny Rules still apply. Manual only stops an Allow Rule or AI judgment from running a write without a person.
  </Accordion>

  <Accordion title="Why was a scheduled Automation refused instead of asking?">
    Automations use **Auto · Rules + AI · Refuse** by default, because no person may be there to answer. Accept a Rule for that write, or set the Automation to an **Ask me** choice.
  </Accordion>

  <Accordion title="What does AI judgment allow?">
    A write that matches what you asked for and stays in its scope. A destructive write or a write with an unclear effect goes to step 3. A destructive write that waits for a person also raises a high-severity alert.
  </Accordion>

  <Accordion title="Can I pick AI judgment for one chat?">
    Yes, when the **Let people pick AI judgment for one chat** switch is on. Otherwise you see "An admin turned off AI judgment in conversations."
  </Accordion>
</AccordionGroup>

## Related

<CardGroup cols={2}>
  <Card title="Approvals" icon="shield-check" href="/guide/approvals/overview">
    See how Rules, the mode, and approvers decide each write
  </Card>

  <Card title="Approval Rules" icon="list-check" href="/guide/approvals/rules">
    Accept Rules so common writes do not depend on the mode
  </Card>

  <Card title="Automations" icon="calendar-check" href="/guide/automation/automations">
    Run agent instructions on schedules, webhooks, and repository events
  </Card>

  <Card title="Autonomous Operations" icon="robot" href="/guide/automation/autonomous-agents">
    See what runs without a live prompt and how the mode bounds it
  </Card>
</CardGroup>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.