Split Lesson 3 into Context (L3) and Custom Instructions (L4)

Lesson 3 covered too much ground. Split it into two focused lessons:
- Lesson 3: Context — how Copilot finds context, @file mentions,
  managing the context window (/clear, /compact, /context), anti-patterns
- Lesson 4: Custom Instructions — copilot-instructions.md, AGENTS.md,
  the golden rule, resist the manual refactor

Renumbered all subsequent lessons (4-22 → 5-23), validators, and updated
every cross-reference across agents, hooks, copilot-instructions.md,
README, and lesson files.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Author: matthewwilson

.github/agents/core.agent.md

@@ -7,7 +7,7 @@ Switch back to the core track. Do the following steps in order:
 
 1. Read the file `tutorial/state.json`
 2. Set the `track` field to `"core"`
-3. If `currentLesson` is greater than 17, set `currentLesson` to 17
+3. If `currentLesson` is greater than 18, set `currentLesson` to 18
 4. Write the updated JSON back to `tutorial/state.json`
 5. Determine the lesson filename: look in `tutorial/lessons/` for the file whose name starts with the zero-padded lesson number (e.g., lesson 17 becomes "17-")
 6. Read that lesson file

.github/agents/next.agent.md

@@ -9,7 +9,7 @@ Advance the tutorial to the next lesson. Do the following steps in order:
 2. Get the current lesson number from the `currentLesson` field
 3. Add the current lesson number to the `completedLessons` array if it is not already there
 4. Increment `currentLesson` by 1
-5. Read the `track` field from state (default `"core"` if missing). If `track` is `"qa"`, cap `currentLesson` at 22. If `track` is `"core"`, cap at 17. (Set it back to the cap value if it exceeds it.)
+5. Read the `track` field from state (default `"core"` if missing). If `track` is `"qa"`, cap `currentLesson` at 23. If `track` is `"core"`, cap at 18. (Set it back to the cap value if it exceeds it.)
 6. Write the updated JSON back to `tutorial/state.json`
 7. Determine the next lesson filename: look in `tutorial/lessons/` for the file whose name starts with the zero-padded lesson number (e.g., lesson 3 becomes "03-", lesson 7 becomes "07-")
 8. Read that lesson file

.github/agents/progress.agent.md

@@ -13,30 +13,31 @@ Show the user their tutorial progress. Do the following:
    - 0: Welcome
    - 1: First Conversation
    - 2: Effective Prompting & Verification
-   - 3: Context & Custom Instructions
-   - 6: Planning Mode
-   - 8: Skills & Custom Agents
-   - 9: Agent Delegation & Collaboration
-   - 10: Hooks in Action
-   - 12: Session Management
-   - 14: Model Selection & Configuration
-   - 15: CLI Tools & MCP Servers
-   - 16: Working with Images
-   - 17: Graduation + Bonus
+   - 3: Context
+   - 4: Custom Instructions
+   - 7: Planning Mode
+   - 9: Skills & Custom Agents
+   - 10: Agent Delegation & Collaboration
+   - 11: Hooks in Action
+   - 13: Session Management
+   - 15: Model Selection & Configuration
+   - 16: CLI Tools & MCP Servers
+   - 17: Working with Images
+   - 18: Graduation + Bonus
 
 **Developer:**
-   - 4: Code Modification
-   - 5: Bug Investigation
-   - 7: Writing & Running Tests
-   - 11: Git Workflow — Commits & PRs
-   - 13: Programmatic Mode & Automation
+   - 5: Code Modification
+   - 6: Bug Investigation
+   - 8: Writing & Running Tests
+   - 12: Git Workflow — Commits & PRs
+   - 14: Programmatic Mode & Automation
 
 4. If `track` is `"qa"`, also show a **QA Track:** section:
-   - 18: Test Planning from Requirements
-   - 19: Test Case Design & Coverage Analysis
-   - 20: Automated Test Generation
-   - 21: E2E Testing with Playwright
-   - 22: QA Track Graduation
+   - 19: Test Planning from Requirements
+   - 20: Test Case Design & Coverage Analysis
+   - 21: Automated Test Generation
+   - 22: E2E Testing with Playwright
+   - 23: QA Track Graduation
 
    If `track` is NOT `"qa"`, show a note instead: "Type /agent qa to start the QA track"
 
@@ -45,4 +46,4 @@ Show the user their tutorial progress. Do the following:
    - `[>]` if it is the `currentLesson` (the one they're on right now)
    - `[~]` if it is in the `skippedLessons` array
    - `[ ]` if it hasn't been reached yet
-6. Show a summary line. If on the core track: "Progress: N of 17 lessons completed". If on the QA track: "Progress: N of 22 lessons completed".
+6. Show a summary line. If on the core track: "Progress: N of 18 lessons completed". If on the QA track: "Progress: N of 23 lessons completed".

.github/agents/qa.agent.md

@@ -3,11 +3,11 @@ description: "Switch to the QA track"
 tools: ["read", "edit"]
 ---
 
-Switch to the QA track (lessons 18-22). Do the following steps in order:
+Switch to the QA track (lessons 19-23). Do the following steps in order:
 
 1. Read the file `tutorial/state.json`
 2. Set the `track` field to `"qa"`
-3. If `currentLesson` is less than 18, set `currentLesson` to 18 (otherwise keep it as-is)
+3. If `currentLesson` is less than 19, set `currentLesson` to 19 (otherwise keep it as-is)
 4. Write the updated JSON back to `tutorial/state.json`
 5. Determine the lesson filename: look in `tutorial/lessons/` for the file whose name starts with the zero-padded lesson number (e.g., lesson 18 becomes "18-")
 6. Read that lesson file

.github/agents/skip.agent.md

@@ -7,7 +7,7 @@ Skip the current exercise and move to the next lesson. Do the following steps:
 
 1. Read `tutorial/state.json`
 2. Add the current lesson number to the `skippedLessons` array if not already there
-3. Increment `currentLesson` by 1. Read the `track` field (default `"core"` if missing). Cap at 22 if `track` is `"qa"`, or at 17 if `track` is `"core"`.
+3. Increment `currentLesson` by 1. Read the `track` field (default `"core"` if missing). Cap at 23 if `track` is `"qa"`, or at 18 if `track` is `"core"`.
 4. Write the updated state back to `tutorial/state.json`
 5. Read the next lesson file from `tutorial/lessons/` (zero-padded prefix)
 6. Present it to the user. Briefly acknowledge the skip in a neutral way — something like "Moving on!" — without making the user feel bad. Everyone learns at their own pace.

.github/copilot-instructions.md

@@ -18,7 +18,7 @@ You are running as an interactive tutorial guide for GitHub Copilot CLI. Your jo
 | `hint` | Get a hint for the current exercise |
 | `skip` | Skip the current exercise |
 | `progress` | See tutorial progress |
-| `qa` | Switch to the QA track (lessons 18-22) |
+| `qa` | Switch to the QA track (lessons 19-23) |
 | `core` | Switch back to the core track |
 
 When suggesting navigation to the user, tell them to type `/agent` and select from the list, or say "next" / "next lesson" / "give me a hint" etc. — you should infer which agent to use from natural language.
@@ -66,26 +66,26 @@ Or, when prompted, they can select **"Yes, approve for session"** to auto-approv
 
 8. **Understand natural language navigation.** When the user says things like "next lesson", "go to the next one", "skip this", "give me a hint", "show my progress", "restart", "switch to QA track" — use the corresponding navigation agent. You don't need to wait for a specific `/agent` command.
 
-## QA Track (Lessons 18-22)
+## QA Track (Lessons 19-23)
 
 The tutorial includes an optional QA track for QA engineers. It is controlled by the `track` field in `tutorial/state.json`:
 
-- `"core"` (default) — Lessons 0-17, the standard developer-focused track
-- `"qa"` — Lessons 18-22, focused on QA workflows
+- `"core"` (default) — Lessons 0-18, the standard developer-focused track
+- `"qa"` — Lessons 19-23, focused on QA workflows
 
 ### QA Track Navigation
 
 | Say this | Or type this | What it does |
 |----------|-------------|-------------|
-| "switch to QA track" | `/agent qa` | Switch to the QA track (starts at lesson 18) |
+| "switch to QA track" | `/agent qa` | Switch to the QA track (starts at lesson 19) |
 | "switch to core track" | `/agent core` | Switch back to the core track |
 
 ### QA Lessons
 
-- **18: Test Planning from Requirements** — Analyze a feature request and create a structured test plan
-- **19: Test Case Design & Coverage** — Identify coverage gaps and design test cases using QA techniques
-- **20: Automated Test Generation** — Turn test cases into running Vitest tests
-- **21: E2E Testing with Playwright** — Write Playwright tests for the web UI
-- **22: QA Track Graduation** — Recap and bonus challenges
+- **19: Test Planning from Requirements** — Analyze a feature request and create a structured test plan
+- **20: Test Case Design & Coverage** — Identify coverage gaps and design test cases using QA techniques
+- **21: Automated Test Generation** — Turn test cases into running Vitest tests
+- **22: E2E Testing with Playwright** — Write Playwright tests for the web UI
+- **23: QA Track Graduation** — Recap and bonus challenges
 
 The `next`, `skip`, and `progress` agents are track-aware — they respect the active track's lesson range. The hooks work with any lesson number automatically.

.github/hooks/validate-exercise.sh

@@ -37,8 +37,8 @@ if [ -f "$VALIDATOR" ]; then
   fi
 fi
 
-# Also run any bonus validators when on the graduation lesson (17)
-if [ "$CURRENT" -eq 17 ]; then
+# Also run any bonus validators when on the graduation lesson (18)
+if [ "$CURRENT" -eq 18 ]; then
   for bonus_validator in "$VALIDATORS_DIR"/validate-bonus-*.sh; do
     [ -f "$bonus_validator" ] || continue
     BONUS_RESULT=$(bash "$bonus_validator" 2>&1) || true

README.md

@@ -52,18 +52,18 @@ You can navigate the tutorial using natural language or the `/agent` command:
 | "hint" or "give me a hint" | `/agent hint` | Get a hint for the current exercise |
 | "skip" or "skip this" | `/agent skip` | Skip the current exercise |
 | "progress" or "show progress" | `/agent progress` | See your tutorial progress |
-| "switch to QA track" | `/agent qa` | Switch to the QA track (lessons 18-22) |
+| "switch to QA track" | `/agent qa` | Switch to the QA track (lessons 19-23) |
 | "switch to core track" | `/agent core` | Switch back to the core track |
 
 ## QA Track
 
-The tutorial includes an optional QA track (lessons 18-22) for QA engineers. It covers test planning, test case design, automated test generation, and E2E testing with Playwright.
+The tutorial includes an optional QA track (lessons 19-23) for QA engineers. It covers test planning, test case design, automated test generation, and E2E testing with Playwright.
 
 To start the QA track, say "switch to QA track" or type `/agent qa`.
 
 To return to the core track, say "switch to core track" or type `/agent core`.
 
-The QA track assumes you have completed at least the core lessons (0-3). The `progress` agent shows both tracks when the QA track is active.
+The QA track assumes you have completed at least the core lessons (0-4). The `progress` agent shows both tracks when the QA track is active.
 
 ## How It Works
 
@@ -73,4 +73,4 @@ The tutorial is built using three Copilot CLI features:
 - **Custom Agents** (`.github/agents/*.agent.md`) — Navigation between lessons
 - **Hooks** (`.github/hooks/`) — State persistence, context injection, and exercise validation
 
-Lesson 10 of the tutorial walks through exactly how these pieces compose together.
+Lesson 11 of the tutorial walks through exactly how these pieces compose together.

tutorial/lessons/00-welcome.md

@@ -18,21 +18,22 @@ You can also use `/agent` and select from the list of tutorial agents.
 
 1. **First Conversation** — Explore a real codebase with Copilot
 2. **Effective Prompting** — Write prompts that get great results on the first try
-3. **Context & Custom Instructions** — Persistent instructions and context management
-4. **Code Modification** — Ask Copilot to write TypeScript code
-5. **Bug Investigation** — Hunt down and fix a real bug
-6. **Planning Mode** — How Copilot plans before it codes
-7. **Writing & Running Tests** — Get Copilot to write and run tests
-8. **Skills & Custom Agents** — Create your own custom agents
-9. **Agent Delegation** — Push work to coding agents and built-in specialists
-10. **Hooks in Action** — The automation that powers this tutorial itself
-11. **Git Workflow** — Commits, PRs, and code review with Copilot
-12. **Session Management** — Resume, compress, and organize your work
-13. **Programmatic Mode** — Run Copilot from scripts and CI/CD
-14. **Model Selection** — Choose the right model for each task
-15. **CLI Tools & MCP Servers** — Use any CLI tool and structured tool access
-16. **Working with Images** — Debug from screenshots and mockups
-17. **Graduation** — Bonus challenges and where to go next
+3. **Context** — How Copilot finds context and how to manage your context window
+4. **Custom Instructions** — Persistent instructions that survive across sessions
+5. **Code Modification** — Ask Copilot to write TypeScript code
+6. **Bug Investigation** — Hunt down and fix a real bug
+7. **Planning Mode** — How Copilot plans before it codes
+8. **Writing & Running Tests** — Get Copilot to write and run tests
+9. **Skills & Custom Agents** — Create your own custom agents
+10. **Agent Delegation** — Push work to coding agents and built-in specialists
+11. **Hooks in Action** — The automation that powers this tutorial itself
+12. **Git Workflow** — Commits, PRs, and code review with Copilot
+13. **Session Management** — Resume, compress, and organize your work
+14. **Programmatic Mode** — Run Copilot from scripts and CI/CD
+15. **Model Selection** — Choose the right model for each task
+16. **CLI Tools & MCP Servers** — Use any CLI tool and structured tool access
+17. **Working with Images** — Debug from screenshots and mockups
+18. **Graduation** — Bonus challenges and where to go next
 
 ## Coming from IDE Copilots?
 
@@ -43,7 +44,7 @@ If you've used GitHub Copilot in VS Code, JetBrains, or other editors, here's ho
 - **Runs commands** — Copilot executes tests, builds, type-checkers, and git commands, then reads the output and reacts. It's not just suggesting code — it's verifying it works.
 - **Persistent memory** — `.github/copilot-instructions.md` files store your conventions, architecture decisions, and rules. Copilot reads them every session — no re-explaining.
 - **Plans before acting** — Plan mode (Shift+Tab) lets Copilot explore and design an approach before writing a single line of code. You review and approve the plan first.
-- **Works with images** — drag-and-drop screenshots, mockups, or error dialogs directly into the conversation (details in Lesson 16).
+- **Works with images** — drag-and-drop screenshots, mockups, or error dialogs directly into the conversation (details in Lesson 17).
 - **Multiple models** — Choose between Claude Sonnet 4.5, Claude Sonnet 4, and GPT-5 with the `/model` command mid-session.
 
 None of this replaces what you already know — it builds on it. If you've been using Copilot for inline completions, chat, or code explanations, Copilot CLI handles all of those and more. This tutorial shows you what's different.