[bot] docs: update Learning Hub with Copilot CLI v1.0.16–1.0.21 features

[github-actions]

What's New

Reviewed CLI changelog for versions released in the past 7 days (v1.0.16–v1.0.21, 2026-04-02 through 2026-04-07) and updated four Learning Hub pages with features not yet documented.

---

New Features Found

Version	Feature
v1.0.21	copilot mcp terminal command for managing MCP servers outside interactive sessions
v1.0.21	Hooks with PascalCase event names now receive VS Code-compatible snake_case payloads (hook_event_name, session_id, ISO 8601 timestamps)
v1.0.20	OpenTelemetry monitoring — copilot help monitoring topic (minor, not yet significant enough for a page)
v1.0.18	Critic agent — auto-reviews plans and complex implementations using a complementary model (experimental, Claude only)
v1.0.18	notification hook event — fires asynchronously on shell completion, permission prompts, elicitation dialogs, and agent completion
v1.0.18	preToolUse permissionDecision response values: "allow" suppresses the tool approval prompt; "ask" raises a PermissionRequest event
v1.0.16	PermissionRequest hook event — programmatically approve or deny tool permission requests
v1.0.17	Built-in skills now ship with the CLI (starting with a guide for customizing the coding agent's cloud environment)

---

Sections Updated

automating-with-hooks.md

• Added notification and PermissionRequest to the Hook Events table
• Added preToolUse permissionDecision output table (allow, ask, non-zero)
• Extended the Cross-Platform Event Name section with the new PascalCase snake_case payload note
• Added a new "Asynchronous Notifications with the notification Hook" practical example showing the async vs. synchronous distinction

understanding-mcp-servers.md

• Added "Managing MCP Servers from the Terminal" section documenting the new gh copilot mcp command and how it complements in-session /mcp slash commands

creating-effective-skills.md

• Added Q&A entry explaining that the CLI now ships built-in skills (v1.0.17+) and what the first built-in skill covers

using-copilot-coding-agent.md

• Added "Critic Agent (Experimental)" section explaining the new auto-review step for plans, how it works, and when it activates

---

Source Announcements

• copilot-cli changelog v1.0.21 — copilot mcp command, PascalCase payload
• copilot-cli changelog v1.0.20 — OpenTelemetry, /yolo persistence
• copilot-cli changelog v1.0.19 — /mcp enable/disable persistence
• copilot-cli changelog v1.0.18 — Critic agent, notification hook, permissionDecision
• copilot-cli changelog v1.0.17 — built-in skills
• copilot-cli changelog v1.0.16 — PermissionRequest hook

Generated by Learning Hub Updater · ◷