Add automations learning hub guide and integrate into docs

- Create new learning hub article: "Using Automations in the GitHub Copilot app"
- Covers practical approach to getting started with automations
- Includes templates, work-surface audit technique, and real examples
- Features the in-app "Awesome Copilot daily PR summary" automation as concrete reference
- Integrates Ashley's Slack guidance on iterative refinement and discovery
- Link from main Learning Hub index and GitHub Copilot app article

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

website/src/content/docs/learning-hub/github-copilot-app.md

@@ -11,6 +11,7 @@ tags:
   - agents
   - parallel-work
 relatedArticles:
+  - ./using-automations-in-copilot-app.md
   - ./using-copilot-coding-agent.md
   - ./agentic-workflows.md
   - ./what-are-agents-skills-instructions.md

website/src/content/docs/learning-hub/index.md

@@ -10,6 +10,8 @@ New to GitHub Copilot? Start here to understand the tools available to you.
 
 **Desktop App**: Explore the [GitHub Copilot app](github-copilot-app/) — a control center for directing multiple agents in parallel. Perfect for agent-native development and parallel work with isolated worktrees.
 
+Ready for practical automation ideas? Start with [Using Automations in the GitHub Copilot app](using-automations-in-copilot-app/) for templates, setup guidance, and real examples.
+
 **Terminal**: Looking for a guided path into GitHub Copilot from the terminal? Explore the [Copilot CLI for Beginners](cli-for-beginners/) with a text-based experience or the [YouTube video series](https://www.youtube.com/watch?v=BDxRhhs36ns&list=PL0lo9MOBetEHvO-spzKBAITkkTqv4RvNl).
 
 ## Fundamentals

website/src/content/docs/learning-hub/using-automations-in-copilot-app.md

@@ -0,0 +1,108 @@
+---
+title: 'Using Automations in the GitHub Copilot app'
+description: 'A practical guide to getting started with Copilot app automations using templates, iterative refinement, and real-world examples.'
+authors:
+  - GitHub Copilot Learning Hub Team
+lastUpdated: 2026-06-17
+estimatedReadingTime: '10 minutes'
+tags:
+  - copilot-app
+  - automations
+  - productivity
+  - workflows
+relatedArticles:
+  - ./github-copilot-app.md
+  - ./agentic-workflows.md
+  - ./using-copilot-coding-agent.md
+prerequisites:
+  - Access to the GitHub Copilot app
+  - A connected repository
+---
+
+Automations are one of the fastest ways to make the GitHub Copilot app useful every day. Instead of manually asking Copilot to do the same recurring work, you save the task once and run it on a schedule or on demand.
+
+If you are not sure where to begin, start with templates, adapt one to your workflow, and iterate from real runs.
+
+## Start with templates (then customize)
+
+When you create a new automation in the Copilot app, browse the built-in templates first. They give you a strong starting point for both prompt structure and scope.
+
+Common examples include:
+
+- **Triage incoming issues** (for example, label issues as `bug`, `enhancement`, or `other`)
+- **Fix failing tests nightly** (attempt a fix and open a draft pull request)
+- **Prepare weekly release notes** (draft and open a pull request on schedule)
+
+Even if none of these are an exact match, templates are the quickest way to avoid a blank-page start.
+
+## Use the work-surface audit trick
+
+A practical way to discover useful automations is to ask Copilot to audit your work surfaces and suggest candidates.
+
+Try this in a regular chat first:
+
+> Look across all of my work surfaces. Browse Teams and Outlook through WorkIQ, use the Slack MCP server to look at Slack, check my calendar, and figure out where I am dropping balls, where I might need help, and suggest what automations would be useful.
+
+This often produces several concrete automation ideas in one pass. Then turn the best one into a saved automation.
+
+## Create your first automation in the app
+
+1. Open **Automations** in the Copilot app sidebar.
+2. Click **New automation**.
+3. Start from a template or from scratch.
+4. Give it a clear name and a specific prompt.
+5. Choose when it runs (manual, hourly, daily, or weekly).
+6. Optionally set mode, model, and reasoning effort.
+7. Use **Create and run** for the first run so you can immediately inspect output and refine.
+
+If your first version is only 70% right, that is normal. The fastest path is to iterate from a real run.
+
+## Example: Awesome Copilot daily PR summary
+
+Here is a real in-app automation used on the `github/awesome-copilot` repository:
+
+| Field | Value |
+|---|---|
+| **Name** | Awesome Copilot daily PR summary |
+| **Interval** | Daily at 09:00 |
+| **Mode** | Autopilot |
+| **What it does** | Pulls open PRs via `gh api`, filters to updates in the last 24 hours, and returns a concise summary table |
+
+Why this works well:
+
+- It is narrowly scoped (one repo, one reporting goal).
+- It has explicit command examples in the prompt.
+- It runs on a predictable cadence and produces a format that is easy to review.
+
+This is a strong starter pattern you can copy for issue triage, release prep, review tracking, or team digests.
+
+## Iterate in chat to improve results
+
+After each run, open a chat and refine the automation prompt directly. Typical refinements:
+
+- Tighten scope (for example, only specific labels, teams, or paths)
+- Improve output format (table, checklist, short summary)
+- Add clear success criteria (what should be included or excluded)
+
+The goal is not a perfect first prompt. The goal is a useful automation that gets better with each run.
+
+## Real-world ideas from the community
+
+Teams are already using recurring automations for practical work such as:
+
+- Daily status updates for initiative-specific issues
+- Review and inbox hygiene tasks
+- Recurring personal workflow maintenance
+
+Use these as patterns, then tailor them to your own repo, rituals, and communication style.
+
+## Practical guardrails
+
+- Keep prompts explicit and scoped to one outcome.
+- Give the automation only the capabilities it needs.
+- Avoid secrets in prompts; use repository secrets and variables where needed.
+- Prefer review-friendly outputs so it is easy to trust and iterate.
+
+## Next step
+
+Open **Automations**, pick one template, and convert one recurring task you currently do manually into a daily run. That single win usually makes the next automation obvious.