Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
Expand Up @@ -49,6 +49,9 @@ The models available for {% data variables.product.prodname_copilot_short %} var
### United States

{% data reusables.copilot.model-compliance.us-models %}
* {% data variables.copilot.gemini_4_carbon %}
* {% data variables.copilot.gemini_4_hydrogen %}
* {% data variables.copilot.gemini_4_neon %}

### European Union

Expand All @@ -63,12 +66,17 @@ The models available for {% data variables.product.prodname_copilot_short %} var
* {% data variables.copilot.gpt_6_luna %}
* {% data variables.copilot.gpt_6_sol %}
* {% data variables.copilot.gpt_61_sol %}
* {% data variables.copilot.gpt_62_astra %}
* {% data variables.copilot.gpt_62_sol %}
* {% data variables.copilot.claude_haiku_45 %}
* {% data variables.copilot.claude_haiku_55 %}
* {% data variables.copilot.claude_opus_55 %}
* {% data variables.copilot.claude_sonnet_5 %}
* {% data variables.copilot.claude_sonnet_55 %}
* {% data variables.copilot.gemini_4_argon %}
* {% data variables.copilot.gemini_4_carbon %}
* {% data variables.copilot.gemini_4_hydrogen %}
* {% data variables.copilot.gemini_4_neon %}

## Pricing changes

Expand Down
1 change: 1 addition & 0 deletions content/admin/troubleshooting-your-instance/index.md
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,7 @@ children:
- /troubleshooting-tls-and-certificates
- /troubleshooting-upgrades-and-migrations
- /troubleshooting-storage-and-disks
- /troubleshooting-github-actions-configuration-and-services
- /troubleshooting-github-actions-data-and-database-health
- /troubleshooting-search-webhooks-and-background-jobs
shortTitle: Troubleshoot your instance
Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,9 @@
---
title: Troubleshooting {% data variables.product.prodname_actions %} configuration and services
intro: 'Resolve problems with the setup and services that run {% data variables.product.prodname_actions %} on {% data variables.location.product_location %}.'
versions:
ghes: '*'
children:
- /stalled-notification-task
shortTitle: Actions configuration
---

Large diffs are not rendered by default.

Original file line number Diff line number Diff line change
Expand Up @@ -106,6 +106,9 @@ After your tests generate a Cobertura XML report, upload it to {% data variables
* **`LABEL`**: A label to identify this coverage report (for example, `code-coverage/pytest`).
1. Commit and push the workflow change.

> [!NOTE]
> The upload action checks that your file is a Cobertura XML report. If it isn't, for example because your test tool produced an LCOV report, the upload step fails with the error `This file is not a Cobertura XML report. Upload a Cobertura XML report.` To fix this, configure your test framework to output Cobertura XML, or convert your report to Cobertura XML before the upload step.

### Full workflow example

This example runs Python tests with `pytest-cov` and uploads the coverage report:
Expand Down
2 changes: 1 addition & 1 deletion content/copilot/concepts/copilot-surfaces/copilot-cli.md
Original file line number Diff line number Diff line change
Expand Up @@ -245,7 +245,7 @@ Typically, you can choose from three options:

**Option 1** allows {% data variables.product.prodname_copilot_short %} to run this particular command, this time only. The next time it needs to use this tool, it will ask you again.

**Option 2** allows {% data variables.product.prodname_copilot_short %} to use this tool again, without asking you for permission, for the duration of the currently running session. It will ask for your approval again in new sessions, or if you resume the current session in the future. If you choose this option, you are allowing {% data variables.product.prodname_copilot_short %} to use this tool in any way it thinks is appropriate. For example, if {% data variables.product.prodname_copilot_short %} asks you to allow it to run the command `rm ./this-file.txt`, and you choose option 2, then {% data variables.product.prodname_copilot_short %} can run any `rm` command (for example, `rm -rf ./*`) during the current run of this session, without asking for your approval.
**Option 2** allows {% data variables.product.prodname_copilot_short %} to use this tool again, without asking you for permission, for the duration of the session. The CLI restores these approvals when you resume the session, so it won't ask again for requests covered by them. It will ask for your approval again in new sessions. Directory approvals and approvals granted automatically by the safety review or a policy aren't restored when you resume a session. If you choose this option, you are allowing {% data variables.product.prodname_copilot_short %} to use this tool in any way it thinks is appropriate. For example, if {% data variables.product.prodname_copilot_short %} asks you to allow it to run the command `rm ./this-file.txt`, and you choose option 2, then {% data variables.product.prodname_copilot_short %} can run any `rm` command (for example, `rm -rf ./*`) during the current run of this session, without asking for your approval.

**Option 3** cancels the proposed command and allows you to tell {% data variables.product.prodname_copilot_short %} to try a different approach.

Expand Down
2 changes: 1 addition & 1 deletion content/copilot/how-tos/copilot-cli/cli-best-practices.md
Original file line number Diff line number Diff line change
Expand Up @@ -235,7 +235,7 @@ For best results on complex tasks:
To view information about the current CLI session, enter:

```copilot
/session
/session info
```

To view a list of any session checkpoints, enter:
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -63,11 +63,11 @@ Install {% data variables.copilot.copilot_cli_short %}. See [AUTOTITLE](/copilot

Allow {% data variables.product.prodname_copilot_short %} to use this tool. The next time {% data variables.product.prodname_copilot_short %} wants to use this tool, it will ask you to approve it again.

**2. Yes, and approve TOOL for the rest of the running session**:
**2. Allow the tool for this location or session**:

Allow {% data variables.product.prodname_copilot_short %} to use this tool—with any options—without asking again, for the rest of the currently running session. Any pending parallel permission requests of the same type will be auto-approved. You will have to approve the command again in future sessions.
The wording and scope depend on the request. An option that says "don't ask again" in this repository or directory saves the approval for that location, including future sessions. An option that says "for the rest of the running session" applies only to the current session.

Choosing this option is useful for many tools—such as `chmod`—as it avoids you having to approve similar commands repeatedly in the same session. However, be aware of the security implications of this option. For example, choosing this option for the command `rm` would allow {% data variables.product.prodname_copilot_short %} to delete any file in the current directory or its subdirectories without asking for your approval.
Saving an approval is useful for many tools—such as `chmod`—as it avoids you having to approve similar commands repeatedly. However, be aware of the security implications of this option. For example, saving an approval for the command `rm` would allow {% data variables.product.prodname_copilot_short %} to delete any file in the current directory or its subdirectories without asking for your approval.

**3. No, and tell Copilot what to do differently (Esc)**:

Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -97,7 +97,7 @@ Below are some example research prompts and explanations of why they work well.

## Viewing and sharing a research report

When the research completes, use the displayed link to open the full report in your default Markdown editor. Alternatively, press <kbd>Ctrl</kbd>+<kbd>Y</kbd> to open the most recent research report from the current session.
When the research completes, use the displayed link to open the full report in your default Markdown editor. Alternatively, if the session has no non-empty plan, press <kbd>Ctrl</kbd>+<kbd>Y</kbd> to open the most recent research report from the current session. If a non-empty plan exists, this shortcut opens the plan instead.

> [!NOTE]
> The editor used by <kbd>Ctrl</kbd>+<kbd>Y</kbd> is determined by the `COPILOT_EDITOR`, `VISUAL`, or `EDITOR` environment variables, in that order. On Linux and macOS, if none of these variables are set, {% data variables.copilot.copilot_cli_short %} uses vi.
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -31,7 +31,7 @@ Rewind works in any directory, including folders that aren't Git repositories. I

Use <kbd>Esc</kbd> when you want gradual, targeted control. While {% data variables.product.prodname_copilot_short %} is working, the first press displays a reminder and a second press interrupts the operation. A single press also closes an open dialog or denies a pending permission request.

Use <kbd>Ctrl</kbd>+<kbd>C</kbd> when you want to cancel without the confirming second press that <kbd>Esc</kbd> requires. A single <kbd>Ctrl</kbd>+<kbd>C</kbd> immediately cancels the active operation and clears any queued prompts. Any file write that is already in progress will complete—files are not left corrupted mid-write—but any remaining planned changes are abandoned. Pressing <kbd>Ctrl</kbd>+<kbd>C</kbd> a second time within two seconds, when the input area is empty, exits the session entirely.
Use <kbd>Ctrl</kbd>+<kbd>C</kbd> when you want to cancel without the confirming second press that <kbd>Esc</kbd> requires. A single <kbd>Ctrl</kbd>+<kbd>C</kbd> immediately cancels the active operation and clears any queued prompts. Any file write that is already in progress will complete—files are not left corrupted mid-write—but any remaining planned changes are abandoned. Once no work is running and the input area is empty, press <kbd>Ctrl</kbd>+<kbd>C</kbd> twice within two seconds to exit.

### What pressing Esc does in different situations

Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -58,6 +58,8 @@ copilot -p "YOUR PROMPT" --max-ai-credits NUMBER

## What happens when the limit is reached

When usage reaches 90% of the limit but remains below it, the next model call is a final response without tools unless you remove the limit or raise it enough to bring usage below 90%. In interactive mode, you are prompted to make this choice before that call.

When the limit is hit, the agent stops cleanly and lets you know.

* **In interactive mode**, you are prompted to reset the limit. You can use `/limits set` to raise the limit and continue your session from where the agent stopped.
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -33,7 +33,7 @@ Typically, you'll use the `/fleet` slash command after creating an implementatio
1. Once the plan is complete, select one of the following options:

* **Accept plan and continue in Autopilot execution mode + /fleet** to allow {% data variables.product.prodname_copilot_short %} to use subagents and work autonomously to implement the plan without any further input.
* **Exit plan mode and I will prompt myself** and then enter a prompt such as `/fleet implement the plan`. {% data variables.product.prodname_copilot_short %} will start working on the plan, using subagents to run parts of the work in parallel where possible. It may ask you to answer questions or make decisions as it works through the plan.
* **Exit to Interactive execution mode without running the plan** and then enter a prompt such as `/fleet implement the plan`. {% data variables.product.prodname_copilot_short %} will start working on the plan, using subagents to run parts of the work in parallel where possible. It may ask you to answer questions or make decisions as it works through the plan.

### Assigning specialized subagents

Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -26,7 +26,7 @@ You can start Autopilot mode from an interactive {% data variables.copilot.copil
```

1. Press <kbd>Shift</kbd>+<kbd>Tab</kbd> until the status line shows Autopilot mode.
1. If the CLI asks how to handle permissions, choose whether to enable all permissions, use manual approval for the session, or cancel.
1. If the CLI asks how to handle permissions, choose whether to enable all permissions, use {% data variables.copilot.assisted_permissions %} or manual approval for the session, or cancel.
1. Enter a specific prompt that includes the outcome and checks you expect. For example:

```copilot
Expand Down Expand Up @@ -64,13 +64,12 @@ This command moves from planning to Autopilot mode automatically when the plan i

## Configuring permissions

When entering autopilot mode, if you have not already granted {% data variables.product.prodname_copilot_short %} all permissions, a message is displayed prompting you to choose between three options:
When entering autopilot mode, if you have not already granted {% data variables.product.prodname_copilot_short %} all permissions, the CLI may prompt you to choose a permission mode:

```text
1. Enable all permissions (recommended)
2. Use Manual Approval for this session
3. Cancel (Esc)
```
1. **Use Allow All for this session** grants broad permissions without reviewing individual actions. If the option also says **disable sandbox for this session**, selecting it disables local sandboxing for the current session. The sandbox remains enabled when organization policy requires it or sandbox bypass is disabled.
1. **Use {% data variables.copilot.assisted_permissions_caps %} for this session** keeps the current sandbox setting and automatically approves requests an AI judges safe. Other requests that need approval are denied in Autopilot mode. This option is not offered when administrator policy disables it.
1. **Use Manual Approval for this session** automatically denies tool requests that require approval.
1. **Cancel** cancels the switch to Autopilot mode.

You will get the best results from Autopilot mode if you enable all permissions. If you choose manual approval, {% data variables.product.prodname_copilot_short %} will automatically deny any tool requests that require approval, which may prevent it from completing certain tasks. You can change your mind later and grant full permissions, during an Autopilot session, by using the `/allow-all` command (or its alias `/yolo`).

Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -167,7 +167,7 @@ Sessions can also run **remotely**. When you delegate a task with [`/delegate`](

### Stopping a session

To stop the session you are currently working in without deleting it, use the `/exit` slash command. The session is saved and you can resume it later.
To stop the session you are currently working in without deleting it, use the `/exit` slash command. By default, if the sidebar is open and another live session is available, only the current session closes. Otherwise, the CLI exits. The session is saved and you can resume it later.

Alternatively, you can permanently delete a session from the sidebar by highlighting it and pressing <kbd>x</kbd> twice. {% data reusables.copilot.close-session %}

Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -1685,6 +1685,12 @@ When a shell command needs tool approval and access to paths outside the allowed

Use `/reset-allowed-tools` to clear session approvals for the current session.

### Session approvals on resume

When you resume a session with `copilot --resume`, the CLI restores approvals you granted for the rest of that session for commands, tools, URLs, parent-domain scopes, and file paths. It also restores session denials, so requests covered by these decisions do not prompt again.

Directory approvals and approvals granted automatically by the safety review or a policy are not restored. Current managed policy still applies.

### Parent-domain URL approvals

URL approval prompts can offer a parent-domain scope, such as `https://*.example.com`, in addition to the exact origin. This approves every subdomain of the registrable domain, such as `api.example.com` and `flow.example.com`, for the rest of the session.
Expand Down Expand Up @@ -1721,6 +1727,8 @@ Shell commands are analyzed before execution to identify potentially dangerous p

High-risk commands can trigger additional warnings and approval prompts. Whether a request needs your approval depends on the permissions in effect.

The Bash tool identifies executables, file write redirections, and possible URLs in command arguments. Read-only commands such as `cat` and `ls` are automatically approved, subject to applicable restrictions. If a script cannot be parsed, the CLI shows a warning and always prompts for approval of the full script. Allow rules and session approvals do not apply to that request.

When sandboxing is enabled and the **Allow sandbox bypass** setting is on (the default), a synchronous shell command that's blocked by the sandbox's filesystem or network policy prompts you to re-run it outside the sandbox—no model round-trip is required. Approving the prompt re-runs the command and returns its output. Declining keeps the sandboxed (blocked) result.

A sandbox bypass prompt also offers a **Yes, and disable the sandbox for the rest of this session** option. Choosing it turns off sandboxing for every remaining command in the current session, not just the command being approved. This only applies to the current session—starting a new session with `/new` sandboxes commands again, and the option has no effect on the saved `sandbox` configuration setting or on other running sessions.
Expand All @@ -1735,6 +1743,14 @@ When a denial is caused by a symlink whose target escapes what the sandbox polic

For more information, see [AUTOTITLE](/copilot/how-tos/cloud-and-local-sandboxes/configuring-local-sandbox-settings#allowing-sandbox-bypass).

#### File and subagent permissions

By default, file access is limited to the current working directory and its subdirectories, and the system temporary directory. File edits require `write` permission or explicit approval, except for session workspace files such as plans and checkpoints, which are automatically approved. A file edit that also needs access to a path outside the allowed directories can request both approvals in a single prompt.

When sandboxing is enabled and bypass is allowed, `apply_patch` and the `view`, `create`, `str_replace`, and `insert` operations of `str_replace_editor` prompt to run outside the sandbox if its filesystem policy denies the path. Approval applies once, and the timeline marks the operation as a sandbox bypass.

Subagents inherit permissions from the parent session and cannot escalate them.

#### Environment variable denylist

The CLI blocks inline assignment of environment variables that can be exploited to execute arbitrary code even in otherwise read-only commands. Blocked categories include:
Expand Down
Loading
Loading