yousufkidiya17/hermes-mobile-modified/youtube-to-rag-pipeline

Claude Code and agent skill by yousufkidiya17/hermes-mobile-modified

Trust tier 2
  • community
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Build a RAG from YouTube transcripts for personality clones.

Install

npx vskill@latest install yousufkidiya17/hermes-mobile-modified/youtube-to-rag-pipeline

SKILL.md

The same SKILL.md was published earlier as yousufkidiya17/hermes-mobile-modified/youtube-to-rag-pipeline.

YouTube Transcript → RAG Pipeline

Build a searchable knowledge base from YouTube transcripts that captures a creator's speaking style, catchphrases, vocabulary, and personality — not just factual content.

When to Use

  • User wants to clone a YouTuber/creator's AI personality
  • Building a RAG that captures tone, style, and catchphrases
  • Learning RAG hands-on with real YouTube data

Pipeline

YouTube URL → Transcript → Chunks → Embeddings → Vector DB → Query → Personality Response

Step 1: Get Transcripts

Use youtube-transcript-api:

bash
pip install youtube-transcript-api
python3 -c "
from youtube_transcript_api import YouTubeTranscriptApi
transcript = YouTubeTranscriptApi.get_transcript('VIDEO_ID')
for t in transcript:
print(f'{t[\"start\"]}: {t[\"text\"]}')
"

Prefer solo content (comedy specials, monologues, solo vlogs) over multi-speaker videos.

Step 2: Chunk & Organize

Split into 200-500 word semantic chunks. Organize by style type:

CategoryWhat It Captures
speaking_styleHinglish mix, sentence rhythm, filler words
catchphrasesSignature phrases, repeated expressions
humor_styleRoast pattern, punchline structure
vocabularyUnique words, slang, domain terms

Step 3: Store in Chroma

python
import chromadb
client = chromadb.PersistentClient(path='./rag_db')
collection = client.create_collection(name='personality')
collection.add(documents=chunks, metadatas=[...], ids=[...])

Step 4: Query

Query for style: "Kaise roast karta hai?", "Signature dialogue kya hai?" The RAG returns relevant style chunks → inform the AI's response.

Step 5: System Prompt Assembly

You are [PERSON]. Speak exactly like them:
- Mix Hindi/English
- Signature phrases: {extracted}
- Humor style: {extracted}

Dependencies

  • youtube-transcript-api
  • chromadb
  • sentence-transformers

Pitfalls

  • Multi-speaker videos need filtering — prefer solo content
  • Transcripts miss delivery tone (sarcasm) — note it in metadata
  • Focus on HOW they speak, not WHAT they say

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
5

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/yousufkidiya17/hermes-mobile-modified/youtube-to-rag-pipeline/badge)](https://verified-skill.com/skills/yousufkidiya17/hermes-mobile-modified/youtube-to-rag-pipeline)

Something wrong with this skill or its page? Tell us.