Auto mode changedclaude-science/auto-mode
Nearest release: v2.1.312, published 2 hours before upstream edited the page. Shown because the two are within 24 hours of each other. Nothing here says the release caused the edit.
Upstream edited this page at 27 Oct 2026 20:48 UTC, give or take a minute or two: the time comes from Anthropic’s own sitemap rather than from a commit. This site recorded the change at 27 Oct 2026 21:07 UTC.
Upstream edited
Recorded here
Lines+18added
Lines−12removed
From line
1
where the diff opens
First seen
22 Oct 2026
this site's first read of the page
Recorded edits3to this page, all time
## What each check sends to Anthropic
The whole hunk
from line 1, old and new numbered
/
from line 1
11# Auto mode
22
3> Auto mode in Claude Science lets a safety check approve code and connector calls.
3> Auto mode lets Claude Science run code and call connectors without stopping to ask about every step.
44
5Auto mode in Claude Science lets Claude run code and call connectors without routine [permission cards](https://claude.com/docs/claude-science/core-concepts.md#permission-cards). A separate safety check reviews code, and the connector calls that would otherwise ask you, before they run. It approves the ones that look safe and holds anything that escalates beyond your request, is destructive or hard to undo, or appears driven by hostile content Claude read. Anything the check holds or can't check still shows you the card. Allow or deny the request as you would with auto mode off.
5Auto mode lets Claude Science run code and call connectors without stopping to ask about every step. Instead, Claude Science reviews these actions for safety (such as checking for data exfiltration or [prompt injection](https://support.claude.com/en/articles/13364135-use-claude-cowork-safely)) and automatically blocks anything it determines to be unsafe. When an action is blocked, Claude Science pauses and asks you directly. If Claude Science keeps running into blocks, it switches back to asking your permission for each step.
66
77> [!WARNING]
88>
9> Auto mode in Claude Science reduces permission cards but does not guarantee safety. Use it for tasks where you trust the general direction, not as a replacement for review on sensitive operations.
9> No defense is perfect and no mode replaces your judgment. For work with real consequences, such as handling important files, stay close and review what Claude Science does, or consider turning auto mode off.
1010
1111## Turn auto mode on or off
1212
13Where auto mode is available to you, it starts on.
13Where available, auto mode is enabled by default for Pro, Max, and Team plans and off by default for Enterprise plans unless your organization's admin allows it.
1414
15* **For new sessions**: go to **Settings > Permissions** and turn the **Auto mode on for new sessions** toggle off or on.
16* **For the current session**: open the session settings menu in the message box and turn the **Auto mode** row on or off.
15* **To set your default for new sessions**: go to **Settings > Permissions** and turn the **Auto mode on for new sessions** toggle off or on.
16* **To turn auto mode on or off for the current session**: open the session settings menu in the message box and turn the **Auto mode** row on or off.
1717
18The **Auto mode on for new sessions** toggle also decides for sessions that don't have their own auto mode setting, such as sessions you started before Anthropic made auto mode available to you. A session with its own setting keeps it when you change the toggle, so turn auto mode off in that session from the session settings menu in the message box.
18The **Auto mode on for new sessions** toggle also turns auto mode on and off in your older sessions, from before you had auto mode. A session with its own setting keeps it until you change the Auto mode row in that session's settings menu.
1919
2020## What auto mode needs in Claude Science
2121
from line 22
2222Auto mode in Claude Science is available only when all of these requirements are met:
2323
2424* **Sandbox**: the [sandbox](https://claude.com/docs/claude-science/core-concepts.md#sandbox) is on. With the sandbox off, auto mode approves nothing.
25* **Organization**: on Team and Enterprise plans, your organization's admin allows it (see [Auto mode in Admin controls](https://claude.com/docs/claude-science/admin-controls.md#auto-mode)).
25* **Organization**: on Team and Enterprise plans, your organization's admin allows it (see [Auto mode in Admin controls](https://claude.com/docs/claude-science/admin-controls.md#auto-mode)). If your role is set to **Custom**, one of your custom roles must also include it.
2626* **Availability**: Anthropic has made auto mode available to you in Claude Science.
2727
2828If the toggle is missing or grayed out, one of these requirements is unmet.
from line 29
2929
3030## What auto mode checks
3131
32Auto mode in Claude Science checks code that Claude runs in the sandbox (Python, R, shell, and PowerShell cells) and calls to remote and directory connectors. A permission you saved for code cells doesn't skip the check while auto mode is on.
32Auto mode in Claude Science checks:
3333
34Everything else asks as it does with auto mode off. That includes network access, access to files outside the session's workspace folder, remote compute, Claude's package and environment tools, and any tool you set to **Ask each time**.
34* Code that runs in the sandbox (Python, R, shell, and PowerShell cells)
35* Code you've already allowed
36* Calls to remote and directory connectors that would otherwise ask you
3537
36## What each check sends to Anthropic
38Even when auto mode is on, Claude Science still asks permission for:
3739
38Each check in Claude Science is a separate request to a Claude model at Anthropic and counts toward your usage. It includes the pending action in full and these parts of the conversation so far: your messages, Claude's questions with your answers, and each action Claude took. Each action still in the conversation is sent in full, such as the code Claude ran, the text it wrote to a file, a short message it sent you while it worked, or the instructions it wrote when it delegated part of the work. Claude's other replies and the output of those actions are left out, except where a summary of older parts of a long conversation restates them. When code runs anywhere but the top level of the Claude Science session's workspace folder, the request includes the full path of the folder the code runs in, which can include your computer's user name.
40* Network access
41* Access to files outside the session's workspace folder
42* Remote compute
43* Claude Science's package and environment tools
44* Any tool you set to **Ask each time**
3945
No line in this hunk matches that.