github · github-actions · Jul 17, 2026
@@ -3,7 +3,7 @@ title: 'Agents and Subagents'
 description: 'Learn how delegated subagents differ from primary agents, when to use them, and how to launch them in VS Code and Copilot CLI.'
 authors:
   - GitHub Copilot Learning Hub Team
-lastUpdated: 2026-07-01
+lastUpdated: 2026-07-17
 estimatedReadingTime: '9 minutes'
 tags:
   - agents
@@ -212,6 +212,12 @@ No. They can run sequentially when one step depends on another, or in parallel w
 
 Yes. In v1.0.66+, usage-based billing users can configure **subagent concurrency and depth limits** directly from `/settings`. The concurrency limit controls how many subagents run in parallel; the depth limit controls how many levels deep delegation can chain (preventing runaway recursive subagent trees). These settings give you predictable control over resource consumption during complex orchestrated tasks.
 
+In v1.0.71, the **default maximum subagent nesting depth was lowered from 6 to 4** to curb runaway recursive delegation. Usage-based billing users can still raise `subagents.maxDepth` up to 128 via `/settings` if their workflows require deeper nesting.
+
+**Can I send follow-up messages to a running subagent?**
+
+Yes, as of v1.0.72, **multi-turn subagents are always enabled**. You can send follow-up messages to a running subagent without waiting for it to finish, allowing you to redirect or clarify mid-task. This was previously a toggle that needed to be enabled.
+
 ## Next steps
 
 - Read [Building Custom Agents](../building-custom-agents/) to design coordinator and worker agents.

@@ -3,7 +3,7 @@ title: 'Copilot Configuration Basics'
 description: 'Learn how to configure GitHub Copilot at user, workspace, and repository levels to optimize your AI-assisted development experience.'
 authors:
   - GitHub Copilot Learning Hub Team
-lastUpdated: 2026-07-13
+lastUpdated: 2026-07-17
 estimatedReadingTime: '10 minutes'
 tags:
   - configuration
@@ -541,16 +541,25 @@ The `/cd` command changes the working directory for the current session. Since v
 
 This is useful when you have multiple backgrounded sessions each focused on a different project directory.
 
-The `/worktree` command (v1.0.61+, also aliased `/move`) creates a new git worktree and switches into it, moving any uncommitted changes along. This lets you start working on a parallel branch without leaving your current terminal session:
+The `/worktree` command (v1.0.61+) creates a new git worktree and switches into it. Since v1.0.71, `/worktree` **leaves your uncommitted changes in the current directory** so you can continue working there while the new worktree starts clean:
 
 ```
 /worktree my-feature-branch
 ```
 
-In v1.0.66+, you can pass a task description to `/worktree` to name the branch from the task and immediately run the task as the first prompt in the new worktree — all in one step:
+The companion `/move` command (v1.0.71+) does the opposite: it **carries your uncommitted changes into the new worktree**, picking up exactly where you left off on a new branch:
 
 ```
-/worktree fix the login redirect
+/move my-feature-branch
+```
+
+> **Before v1.0.71**: `/worktree` and `/move` were aliases for the same command, which always moved uncommitted changes to the new worktree. If you relied on the old behavior of `/worktree` moving your changes, switch to `/move`.
+
+In v1.0.66+, you can pass a task description to either command to name the branch from the task and immediately run the task as the first prompt in the new worktree — all in one step:
+
+```
+/worktree fix the login redirect   # start clean on a new branch
+/move fix the login redirect       # bring uncommitted changes to a new branch
 ```
 
 This creates a branch named from your task description and begins working on it immediately, making it easy to spin up parallel work without stopping to think of a branch name.
@@ -615,6 +624,12 @@ The `/chronicle skills review` subcommand *(v1.0.66+)* opens an interactive revi
 
 This keeps you in control of skill evolution — the agent can propose skill improvements as it discovers reusable patterns, but nothing is applied until you explicitly approve each change.
 
+The `/chronicle cost-tips` subcommand *(v1.0.71+)* opens an interactive cost analysis with recommendations for reducing your token usage. It shows local and cloud cost profiles and suggests changes — such as switching to a lighter model or using `/compact` more frequently — based on patterns in your current session:
+
+```
+/chronicle cost-tips
+```
+
 > **Note**: Session history, file tracking, and the `/chronicle` command were previously experimental features. As of v1.0.40, they are available to all users without enabling experimental mode.
 
 The `/diagnose` command (v1.0.64+) analyzes the current session's logs and surfaces diagnostic information to help troubleshoot unexpected behavior, performance issues, or errors:
@@ -625,6 +640,14 @@ The `/diagnose` command (v1.0.64+) analyzes the current session's logs and surfa
 
 Use `/diagnose` when a session is behaving unexpectedly — it inspects session logs and reports what it finds, making it easier to share diagnostics with support or understand what happened internally.
 
+The `/voice devices` command *(v1.0.71-2+)* opens a picker to choose which microphone the CLI uses for voice mode, and persists that choice across sessions:
+
+```
+/voice devices
+```
+
+Use this if you have multiple microphones (e.g., headset and built-in) and want to pin the CLI to the preferred device without reconfiguring every session.
+
 **Keyboard shortcuts for queuing messages**: Use **Ctrl+Q** or **Ctrl+Enter** to queue a message (send it while the agent is still working). **Ctrl+D** no longer queues messages — it now has its default terminal behavior. If you have muscle memory for Ctrl+D queuing, switch to Ctrl+Q.
 
 **Background running tasks**: Press **Ctrl+X → B** to move the current running task or shell command to the background. The task continues executing while you can type a new message or review earlier output. This is useful for long-running commands where you want to interact with the agent while waiting for the result.
@@ -744,6 +767,8 @@ copilot --plan          # start in plan mode (propose without executing)
 
 This is useful in scripts or CI pipelines where you want the CLI to immediately begin working in a specific mode without an interactive prompt.
 
+> **Note (v1.0.71-2+)**: Plan mode now **hard-blocks** built-in tool calls that would modify the workspace. The agent cannot edit files or run mutating shell commands while planning. Built-in mutators like opening a pull request are also blocked. MCP and external tools are still allowed in plan mode.
+
 The `--max-autopilot-continues` flag controls how many times Copilot can automatically continue in autopilot mode before pausing for confirmation. The default is 5:
 
 ```bash

@@ -3,7 +3,7 @@ title: 'Installing and Using Plugins'
 description: 'Learn how to find, install, and manage plugins that extend GitHub Copilot CLI with reusable agents, skills, hooks, and integrations.'
 authors:
   - GitHub Copilot Learning Hub Team
-lastUpdated: 2026-07-13
+lastUpdated: 2026-07-17
 estimatedReadingTime: '8 minutes'
 tags:
   - plugins
@@ -220,6 +220,16 @@ copilot plugin marketplace update
 
 # Remove a plugin
 copilot plugin uninstall my-plugin
+
+# Remove a specific skill from a plugin (v1.0.72+)
+copilot plugins remove --skill my-skill-name
+```
+
+*(v1.0.72+)* The `plugins` command also supports `--plugin`, `--mcp`, and `--skill` flags for targeted mutations — adding or removing specific components of an installed plugin without reinstalling the whole package:
+
+```bash
+copilot plugins remove --skill slow-skill     # remove just a skill
+copilot plugins remove --mcp unused-server    # remove an MCP server from a plugin
 ```
 
 ### Loading Plugins from a Local Directory

@@ -3,7 +3,7 @@ title: 'Understanding MCP Servers'
 description: 'Learn how Model Context Protocol servers extend GitHub Copilot with access to external tools, databases, and APIs.'
 authors:
   - GitHub Copilot Learning Hub Team
-lastUpdated: 2026-07-13
+lastUpdated: 2026-07-17
 estimatedReadingTime: '8 minutes'
 tags:
   - mcp
@@ -317,6 +317,17 @@ You can also open the `/mcp` manager while the agent is working to toggle server
 
 **Toggling servers on and off** (v1.0.66+): From the `/mcp` list view, you can **enable or disable individual MCP servers** without editing your config file. Select a server in the list and toggle it — disabled servers won't start in future sessions and their tools won't be available to agents. This is useful for temporarily disabling a server that's causing slowdowns or errors without removing it from your configuration entirely.
 
+**Persisting GitHub MCP toolset and tool configuration** *(v1.0.71+)*: You can now persist your GitHub MCP server's toolset and tool selection via `settings.json`. This keeps your GitHub MCP configuration consistent across sessions and repositories:
+
+```json
+{
+  "githubMcpToolsets": ["repos", "issues", "pull_requests"],
+  "githubMcpTools": ["list_issues", "create_pull_request"]
+}
+```
+
+Use `githubMcpToolsets` to enable or disable entire capability groups, and `githubMcpTools` for fine-grained tool-level control. You can also use `--add-github-mcp-tool "*"` at startup to enable all GitHub MCP tools for a session.
+
 **Common causes and fixes**:
 
 | Symptom | Likely Cause | Fix |

@@ -3,7 +3,7 @@ title: 'Working with Canvas Extensions'
 description: 'Create and iterate on GitHub Copilot app canvases using /create-canvas, then shape them into reusable project or personal extensions.'
 authors:
   - GitHub Copilot Learning Hub Team
-lastUpdated: 2026-06-17
+lastUpdated: 2026-07-17
 estimatedReadingTime: '8 minutes'
 tags:
   - copilot-app
@@ -133,6 +133,12 @@ Treat the first `/create-canvas` result as version one. Then refine in-place:
 
 The fastest loop is: **use the canvas**, note friction, and ask the agent for a targeted update.
 
+## Canvas extensions in Copilot CLI
+
+*(v1.0.71-2+)*: Canvas extensions are now supported in the **Copilot CLI** as well as the Copilot app. Extension-driven canvas interactions — where an `extension.mjs` exposes capabilities and a UI panel — work in the CLI's side-panel view.
+
+This means extensions you create for the Copilot app can also be used from the terminal, making canvas workflows portable across surfaces. When an extension provides a canvas, it appears in the CLI's split-pane view, and agents can call its capabilities in the same way as in the app.
+
 ## Next steps
 
 - Review the [GitHub Copilot app overview](../github-copilot-app/) for broader session and workflow concepts.