harut
Ask Harut questions about growing your business to 100k/month. RAG-powered advisor trained on Harut's full 21-hour teachings, personalized for your content creator business (YouTube, Skool, social media). Answers are grounded in the transcript and applied specifically to your situation. Triggers on: ask harut, harut, what does harut say, harut advice, harut says, business growth, 100k month, how do I grow, grow my skool, grow my business, pricing advice, harut question.
소스 정보
- 저장소
- tylerprogramming/claude-skills
- 최근 소스 활동
- 2026년 8월 9일 15:56
- 감지된 SKILL.md 언어
- 영어
- 스타
- 0
- 포크
- 2
설치 방법
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
소스 파일 검토
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
파일 탐색기
3 개 파일SKILL.md 표시 중
SKILL.md
소스 지침 · 읽기 전용 미리보기- name
- harut
- description
- Ask Harut questions about growing your business to 100k/month. RAG-powered advisor trained on Harut's full 21-hour teachings, personalized for your content creator business (YouTube, Skool, social media). Answers are grounded in the transcript and applied specifically to your situation. Triggers on: ask harut, harut, what does harut say, harut advice, harut says, business growth, 100k month, how do I grow, grow my skool, grow my business, pricing advice, harut question.
# Harut - Business Growth RAG Advisor
Harut is your personal business advisor, trained on Harut's 21-hour masterclass on growing a business to $100k/month. It uses RAG (retrieval-augmented generation) to pull the most relevant teachings and applies them to your specific situation: YouTube content, Skool community, and social media.
## Setup (one-time, auto-runs on first query)
The transcript must be indexed before querying. The skill handles this automatically.
## Steps
1. **Check if the index exists:**
```bash
python3 ~/.claude/skills/harut/harut_rag.py status
```
2. **If not indexed yet**, index the transcript:
```bash
python3 ~/.claude/skills/harut/harut_rag.py index ~/content/transcripts/transcript_u2hmXbhTTLE.txt
```
This takes a few minutes and only needs to run once. It will print progress.
3. **Answer the user's question:**
```bash
python3 ~/.claude/skills/harut/harut_rag.py query "QUESTION_HERE"
```
Where QUESTION_HERE is the user's question verbatim.
4. **Present the answer** directly to the user. The answer is already formatted — just show it.
## Notes
- If the transcript file doesn't exist yet, tell the user: "The Harut transcript is still being processed. Run `/transcribe https://youtu.be/u2hmXbhTTLE` first, then try again."
- Index lives at `~/.claude/skills/harut/harut_index.json` — persistent across sessions
- Re-indexing is only needed if the transcript changes
GitHub에서 보기