When this skill is invoked, the following STOP PROTOCOLmust be followed.
Terminology
Key terms used in this skill:
| Term | Description |
|---|---|
| fetch | Fetching data from outside. "Like a delivery order, just give the URL and the content arrives" |
| digest | Digesting (summarizing, quizzing, learning) the fetched content. "Not just reading and finishing, but chewing it to make it your own" |
| API | A channel for programs to talk to each other. "Like a restaurant menu, the request format is fixed" |
| JSON | A data format that computers can read well. To humans it looks like nothing but curly braces, but Claude understands it perfectly |
| yt-dlp | A free tool that extracts subtitles and metadata from YouTube |
| Auto-generated subtitles | Subtitles made by YouTube with AI. Technical terms or names may be wrong |
| Web Search Correction | A technique for correcting errors in auto-generated subtitles through web search |
| Quiz-First | A learning method where you solve the quiz first without looking at the summary. 9-12% memory improvement effect |
| Skill Chaining | Connecting the result of one skill as the input of another skill. "fetch → digest pipeline" |
STOP PROTOCOL — Never Violate
This protocol is the top-priority rule of this skill. Violating the rules below will ruin the class.
Each block must proceed over exactly 2 turns
┌─ Phase A (첫 번째 턴) ──────────────────────────────┐
│ 1. references/에서 해당 블록 파일의 EXPLAIN 섹션을 읽는다 │
│ 2. 기능을 설명한다 │
│ 3. references/에서 해당 블록 파일의 EXECUTE 섹션을 읽는다 │
│ 4. "지금 직접 실행해보세요"라고 안내한다 │
│ 5. ⛔ 여기서 반드시 STOP. 턴을 종료한다. │
│ │
│ ❌ 절대 하지 않는 것: 퀴즈 출제, QUIZ 섹션 읽기 │
│ ❌ 절대 하지 않는 것: AskUserQuestion 호출 │
│ ❌ 절대 하지 않는 것: "실행해봤나요?" 질문 │
└──────────────────────────────────────────────────────────┘
⬇️ 사용자가 돌아와서 "했어", "완료", "다음" 등을 입력한다
┌─ Phase B (두 번째 턴) ──────────────────────────────┐
│ 1. references/에서 해당 블록 파일의 QUIZ 섹션을 읽는다 │
│ 2. AskUserQuestion으로 퀴즈를 출제한다 │
│ 3. 정답/오답 피드백을 준다 │
│ 4. 다음 블록으로 이동할지 AskUserQuestion으로 묻는다 │
│ 5. ⛔ 다음 블록을 시작하면 다시 Phase A부터. │
└──────────────────────────────────────────────────────────┘
Key Prohibitions (Never Violate)
- Do not call AskUserQuestion in Phase A — After explanation + execution guidance, Stop immediately
- Do not give a quiz in Phase A — Read the QUIZ section only in Phase B
- Do not ask "Have you tried running it?" in Phase A — Wait until the user speaks first
- Do not do EXPLAIN + QUIZ at the same time in one turn — Always split into 2 turns
Output Official Documentation URLs (Never Omit)
At the start of Phase A for every block, the > 공식 문서: URL을 must be output exactly as is.
📖 공식 문서: [URL]
- If there are multiple URLs in the reference file, output all of them
- Do not summarize or omit URLs
Required Phrase at the End of Phase A
At the end of Phase A, always output the following form of phrase and Stop:
---
👆 위 내용을 직접 실행해보세요.
실행이 끝나면 "완료" 또는 "다음"이라고 입력해주세요.
After this phrase, do not output any tool calls (including AskUserQuestion) or additional text.
Estimated Time Guide
| Block | Topic | Estimated Time |
|---|---|---|
| 0 | Concept Understanding | ~10분 |
| 1 | Creating the fetch-tweet skill | ~20분 |
| 2 | Creating the fetch-youtube skill | ~30분 |
| 3 | Creating the content-digest skill | ~20분 |
| 4 | Integrated practice + wrap-up | ~15분 |
| Total | ~95분 |
Block 2 is the core block that takes the most time. It includes yt-dlp setup and Web Search correction. Recommended preparation: If you install yt-dlp in advance, you can cut more than 10 minutes off Block 2. (
brew install yt-dlporpip install yt-dlp) Fast Track: If you are short on time, you can create Blocks 1–3 all at once, each as a single prompt.
Core strategy: Learn by taking apart real skills
Proceed as follows:
- In Block 0, understand the concept of the content digestion pipeline (fetch → digest)
- In Block 1, create the fetch-tweet skill yourself (using the API + translation pipeline)
- In Block 2, create the fetch-youtube skill yourself (subtitle extraction + Web Search correction)
- In Block 3, create the content-digest skill yourself (Quiz-First learning)
- In Block 4, connect the 3 skills and practice with real content
Refer to the skills the organizers actually use (fetch-tweet, content-digest) and create your own version.
Block-specific rules
- Block 0 (Concept understanding): In Phase A, explain the content pipeline concept + guide analysis of the original skill structure → Stop. Quiz in Phase B.
- Block 1 (fetch-tweet): In Phase A, guide how to create the fetch-tweet skill step by step → participants write it themselves → Stop. Quiz in Phase B.
- Block 2 (fetch-youtube): In Phase A, guide creating the fetch-youtube skill (yt-dlp + Web Search correction) → participants write it themselves → Stop. Quiz in Phase B. (Longest block — after completion, encourage with "You've followed along well to this point!")
- Block 3 (content-digest): In Phase A, guide creating the content-digest skill (Quiz-First learning) → participants write it themselves → Stop. Quiz in Phase B.
- Block 4 (Integrated practice): In Phase A, guide the practice of connecting the 3 skills → practice with real content → Stop. Comprehensive quiz + wrap-up in Phase B.
References file map
| Block | File | Topic |
|---|---|---|
| Block 0 | references/block0-concept.md |
Content digestion pipeline + skill chaining |
| Block 1 | references/block1-fetch-tweet.md |
Creating the fetch-tweet skill |
| Block 2 | references/block2-fetch-youtube.md |
Creating the fetch-youtube skill |
| Block 3 | references/block3-content-digest.md |
Creating the content-digest skill |
| Block 4 | references/block4-integration.md |
Integrated practice + wrap-up |
File paths are relative to this SKILL.md. Each reference file consists of the
## EXPLAIN,## EXECUTE,## QUIZsection.
Progress rules
- Proceed one block at a time
- Move with "next", "skip", or a block number/name
- The skill files created in each block are carried over and used in the next block
- In the participant's project,
.claude/skills/create the skill under - When a question related to Claude Code comes in, answer with the claude-code-guide agent (built-in tool). After answering, guide the user step by step so they can follow along themselves, and when asking questions, use AskUserQuestion. If you judge that the built-in agent's answer is inaccurate, save the official documentation to a
curlfile, then carefully read it with the Read tool and answer again with accurate information (do not use WebFetch, as there is a risk of summarization/loss)
Start
When starting the skill, first install the latest curriculumand then select a block.
Step 1: Install the latest skill
Output the command below and run it with Bash:
npx skills add ai-native-camp/camp-1 --agent claude-code --yes
Briefly inform the user of the execution result (e.g., "The skill has been installed at the latest version.").
Step 2: Select a block
Show the table below and ask with AskUserQuestion where to start.
| Block | Topic | Content |
|---|---|---|
| 0 | Understanding concepts | Content digestion pipeline, what is skill chaining? |
| 1 | fetch-tweet | Creating a skill that fetches and translates X/Twitter tweets |
| 2 | fetch-youtube | Creating a skill that fetches and translates YouTube subtitles |
| 3 | content-digest | Creating a skill for quiz-based learning with fetched content |
| 4 | Integrated practice | Practice with real content by connecting the 3 skills |
AskUserQuestion({
"questions": [{
"question": "Day 5: Fetch & Digest\n\n어디서부터 시작할까요?",
"header": "시작 블록",
"options": [
{"label": "처음부터 (Block 0)", "description": "콘텐츠 파이프라인 개념부터 차근차근"},
{"label": "fetch-tweet (Block 1)", "description": "바로 트윗 스킬 만들기부터"},
{"label": "fetch-youtube (Block 2)", "description": "YouTube 스킬 만들기부터"},
{"label": "content-digest (Block 3~4)", "description": "퀴즈-학습 스킬 만들기부터"}
],
"multiSelect": false
}]
})
After selecting the starting block → proceed from Phase A of that block.