learning-hub: document CLI v1.0.76 features (Sessions sidebar, /limits predict, MCP snapshots, enterprise sandbox floor)

- copilot-configuration-basics.md: add experimental Sessions sidebar (v1.0.76+), /limits predict command (v1.0.76+), and enterprise sandbox policy floor (v1.0.76+) - understanding-mcp-servers.md: add faster MCP tool loading from snapshots (v1.0.76+) - agents-and-subagents.md: add Sessions sidebar section as complement to /fleet for parallel CLI work Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
2026-08-01 23:12:29 +00:00 · 2026-07-29 21:05:03 +00:00
parent 8ae5a99109
commit 5130e9f0dd
3 changed files with 40 additions and 3 deletions
@@ -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: 2026-07-29
 estimatedReadingTime: '9 minutes'
 tags:
  - agents
@@ -134,6 +134,19 @@ The important behavior is different from a single chat turn:

 That makes `/fleet` a practical way to launch subagents even if you are not authoring custom agent files yourself.

+### Sessions sidebar for parallel CLI work
+
+The experimental **Sessions sidebar** (v1.0.76+) complements `/fleet` by giving you a visual way to manage multiple concurrent sessions side-by-side in the terminal. While `/fleet` handles orchestration automatically within a single session, the Sessions sidebar lets you manually run independent work in separate sessions and switch focus between them without leaving the CLI:
+
+```
+/experimental       # enable experimental features
+/sessions           # open the Sessions sidebar
+```
+
+This is useful when you want to run one session on an open-ended exploration task while another handles a focused implementation — keeping each session's context clean without merging unrelated work.
+
+See [Copilot Configuration Basics](../copilot-configuration-basics/) for full details on the Sessions sidebar and session management commands.
+
 ### Rubber-duck agent

 Available in `/experimental` (v1.0.42+), the **rubber-duck agent** applies a novel multi-model pattern: when you're working in a GPT-powered session, the rubber-duck agent internally routes certain requests through Claude to provide a second perspective. The idea is similar to rubber-duck debugging — talking through a problem with a different "listener" often surfaces assumptions or blind spots you didn't notice.
@@ -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: 2026-07-29
 estimatedReadingTime: '10 minutes'
 tags:
  - configuration
@@ -507,6 +507,17 @@ You can also press **x** on a highlighted session in the session picker (`--resu

 In the session picker, press **`s`** to cycle the sort order: relevance, last used, created, or name. The picker also shows the branch name and idle/in-use status for each session.

+**Experimental Sessions sidebar** (v1.0.76+): An opt-in **Sessions sidebar** lets you run and manage multiple concurrent CLI sessions side-by-side in a split-view layout, without needing to switch terminal windows. Enable it from `/experimental` and open it with `/sessions`:
+
+```
+/experimental       # toggle experimental features, including Sessions sidebar
+/sessions           # open the Sessions sidebar
+```
+
+When the sidebar is active, each session appears as its own panel. You can dispatch new sessions, monitor active ones, and switch focus between them — all without leaving the CLI. This is especially useful when you want to run a long agentic task in the background while continuing interactive work in a separate session.
+
+> **Note**: This is an experimental feature and its behavior may change. Provide feedback via `/feedback` if you find it useful.
+
 The `/rewind` command opens a timeline picker that lets you roll back the conversation to any earlier point in history, reverting both the conversation and any file changes made after that point. You can also trigger it by pressing **double-Esc**:

 ```
@@ -665,6 +676,15 @@ The `/usage` command displays session metrics such as the number of tokens consu
 /usage
 ```

+The `/limits predict` command (v1.0.76+) estimates how much of your usage quota a described task is likely to consume **before you run it**. This is useful when working on large tasks — refactoring a big codebase, running `/fleet` across many files, or long-running agent workflows — where you want to check whether you have sufficient quota headroom:
+
+```
+/limits predict "Refactor the authentication module across all services"
+/limits predict "Run a security audit on the entire codebase"
+```
+
+Copilot returns a token and request estimate for the described task based on the selected model and your current context. Use this to decide whether to proceed immediately, switch to a more efficient model, or compact context first.
+
 The `/compact` command summarizes the conversation history to free up context window space while preserving the thread of the conversation. Use it when your context is getting full but you do not want to start a fresh session:

 ```
@@ -761,6 +781,8 @@ copilot --no-sandbox -p "Set up development environment with system tools"

 These flags apply only to the current invocation — your persisted sandbox preference remains unchanged.

+**Enterprise sandbox policy floor** (v1.0.76+): In enterprise environments, administrators can enforce a **sandbox policy floor** — a minimum sandbox level that applies to all sessions in that organization. When a policy floor is active, the CLI displays a visible indicator showing the enforced minimum, and users cannot disable the sandbox below that floor. The `--no-sandbox` flag and the sandbox toggle in `/settings` are restricted accordingly. This gives organizations predictable security guarantees without requiring per-user configuration.
+
 The `--attachment` flag (available in prompt mode, `-p`) lets you attach files — images or native documents — to the initial prompt in non-interactive mode:

 ```bash
@@ -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: 2026-07-29
 estimatedReadingTime: '8 minutes'
 tags:
  - mcp
@@ -317,6 +317,8 @@ 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.

+**Faster MCP tool loading from snapshots** (v1.0.76+): Copilot CLI now caches a snapshot of each MCP server's tool manifest after the first successful load. On subsequent startups, tools are pre-populated from the snapshot rather than waiting for the server to fully initialize. This significantly reduces the time between launching the CLI and having MCP tools available — particularly noticeable when you have multiple or slow-starting MCP servers. The snapshot is refreshed automatically when the server's tool list changes.
+
 **Common causes and fixes**:

 | Symptom | Likely Cause | Fix |