Angie Jones
162 lines
6.3 KiB
Markdown
162 lines
6.3 KiB
Markdown
---
|
||
sidebar_position: 20
|
||
title: goose Permission Modes
|
||
sidebar_label: goose Permissions
|
||
---
|
||
|
||
import Tabs from '@theme/Tabs';
|
||
import TabItem from '@theme/TabItem';
|
||
import { PanelLeft, Tornado } from 'lucide-react';
|
||
|
||
goose’s permissions determine how much autonomy it has when modifying files, using extensions, and performing automated actions. By selecting a permission mode, you have full control over how goose interacts with your development environment.
|
||
|
||
<details>
|
||
<summary>Permission Modes Video Walkthrough</summary>
|
||
<iframe
|
||
class="aspect-ratio"
|
||
src="https://www.youtube.com/embed/bMVFFnPS_Uk"
|
||
title="goose Permission Modes Explained"
|
||
frameBorder="0"
|
||
allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture"
|
||
allowFullScreen
|
||
></iframe>
|
||
</details>
|
||
|
||
## Permission Modes
|
||
|
||
| Mode | Description | Best For |
|
||
|------|-------------|----------|
|
||
| **Completely Autonomous** | goose can modify files, use extensions, and delete files **without requiring approval** | Users who want **full automation** and seamless integration into their workflow |
|
||
| **Manual Approval** | goose **asks for confirmation** before using any tools or extensions (supports granular [tool permissions](/docs/guides/managing-tools/tool-permissions)) | Users who want to **review and approve** every change and tool usage |
|
||
| **Smart Approval** | goose uses a risk-based approach to **automatically approve low-risk actions** and **flag others** for approval (supports granular [tool permissions](/docs/guides/managing-tools/tool-permissions)) | Users who want a **balanced mix of autonomy and oversight** based on the action’s impact |
|
||
| **Chat Only** | goose **only engages in chat**, with no extension use or file modifications | Users who prefer a **conversational AI experience** for analysis, writing, and reasoning tasks without automation |
|
||
|
||
:::warning
|
||
`Autonomous Mode` is applied by default.
|
||
:::
|
||
|
||
## Configuring goose mode
|
||
|
||
Here's how to configure:
|
||
|
||
<Tabs groupId="interface">
|
||
<TabItem value="ui" label="goose Desktop" default>
|
||
|
||
You can change modes before or during a session and it will take effect immediately.
|
||
|
||
<Tabs groupId="method">
|
||
<TabItem value="session" label="In Session" default>
|
||
|
||
Click the <Tornado className="inline" size={16} /> mode button from the bottom menu.
|
||
</TabItem>
|
||
<TabItem value="settings" label="From Settings">
|
||
1. Click the <PanelLeft className="inline" size={16} /> button on the top-left to open the sidebar.
|
||
2. Click the `Settings` button on the sidebar.
|
||
3. Click `Chat`.
|
||
4. Under `Mode`, choose the mode you'd like.
|
||
</TabItem>
|
||
</Tabs>
|
||
</TabItem>
|
||
<TabItem value="cli" label="goose CLI">
|
||
|
||
<Tabs groupId="method">
|
||
<TabItem value="session" label="In Session" default>
|
||
To change modes mid-session, use the `/mode` command.
|
||
|
||
* Autonomous: `/mode auto`
|
||
* Smart Approve: `/mode smart_approve`
|
||
* Approve: `/mode approve`
|
||
* Chat: `/mode chat`
|
||
</TabItem>
|
||
<TabItem value="settings" label="From Settings">
|
||
1. Run the following command:
|
||
|
||
```sh
|
||
goose configure
|
||
```
|
||
|
||
2. Select `goose settings` from the menu and press Enter.
|
||
|
||
```sh
|
||
┌ goose-configure
|
||
│
|
||
◆ What would you like to configure?
|
||
| ○ Configure Providers
|
||
| ○ Add Extension
|
||
| ○ Toggle Extensions
|
||
| ○ Remove Extension
|
||
// highlight-start
|
||
| ● goose settings (Set the goose mode, Tool Output, Tool Permissions, Experiment, goose recipe github repo and more)
|
||
// highlight-end
|
||
└
|
||
```
|
||
|
||
3. Choose `goose mode` from the menu and press Enter.
|
||
|
||
```sh
|
||
┌ goose-configure
|
||
│
|
||
◇ What would you like to configure?
|
||
│ goose settings
|
||
│
|
||
◆ What setting would you like to configure?
|
||
// highlight-start
|
||
│ ● goose mode (Configure goose mode)
|
||
// highlight-end
|
||
│ ○ Router Tool Selection Strategy
|
||
│ ○ Tool Permission
|
||
│ ○ Tool Output
|
||
│ ○ Max Turns
|
||
│ ○ Toggle Experiment
|
||
│ ○ goose recipe github repo
|
||
│ ○ Scheduler Type
|
||
└
|
||
```
|
||
|
||
4. Choose the goose mode you would like to configure.
|
||
|
||
```sh
|
||
┌ goose-configure
|
||
│
|
||
◇ What would you like to configure?
|
||
│ goose settings
|
||
│
|
||
◇ What setting would you like to configure?
|
||
│ goose mode
|
||
│
|
||
◆ Which goose mode would you like to configure?
|
||
// highlight-start
|
||
│ ● Auto Mode (Full file modification, extension usage, edit, create and delete files freely)
|
||
// highlight-end
|
||
| ○ Approve Mode
|
||
| ○ Smart Approve Mode
|
||
| ○ Chat Mode
|
||
|
|
||
└ Set to Auto Mode - full file modification enabled
|
||
```
|
||
</TabItem>
|
||
</Tabs>
|
||
</TabItem>
|
||
</Tabs>
|
||
|
||
:::info
|
||
In manual and smart approval modes, you will see "Allow" and "Deny" buttons in your session windows during tool calls.
|
||
goose will only ask for permission for tools that it deems are 'write' tools, e.g. any 'text editor write', 'text editor edit', 'bash - rm, cp, mv' commands.
|
||
|
||
Read/write approval makes best effort attempt at classifying read or write tools. This is interpreted by your LLM provider.
|
||
:::
|
||
|
||
## CLI Provider Permission Integration
|
||
|
||
When using [CLI providers](/docs/guides/cli-providers) like Claude Code, goose integrates with the provider's native permission system. In approve mode, permission requests from Claude Code are routed through goose's confirmation interface, giving you a unified experience.
|
||
|
||
For example, with Claude Code in approve mode:
|
||
- Claude Code detects sensitive operations (file writes, shell commands, tool calls)
|
||
- The permission prompt appears in goose's interface (CLI or Desktop)
|
||
- Your allow/deny decision is sent back to Claude Code
|
||
- Claude Code proceeds or adapts based on your response
|
||
|
||
This integration uses the same mechanism as the official Claude Agent SDKs, ensuring compatibility and consistent behavior.
|
||
|
||
See [CLI Providers - Claude Code Configuration](/docs/guides/cli-providers#claude-code-configuration) for setup details.
|