rav4nn

rav4nn / youtube-rag-scraper

Public

Scrape YouTube videos, extract transcripts, and build a semantic search AI knowledge base using RAG and FAISS.

20
2
100% credibility
Found Mar 06, 2026 at 20 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

This repository provides a Python-based command-line tool for scraping metadata and transcripts from YouTube channels, playlists, or videos, generating chunked datasets for retrieval-augmented generation, building vector search indexes, and performing semantic queries.

How It Works

1
📺 Discover the Tool

You hear about a handy program that turns YouTube channels into your own searchable library of videos.

2
🛠️ Get It Ready

You follow easy steps to set up the program on your computer so it can connect to YouTube.

3
Pick Your Focus
👥
Whole Channel

Gather everything from a creator's videos to build a full knowledge collection.

📋
Playlist

Collect from a curated list of videos on a topic.

🎥
Single Video

Pull details from one specific video you care about.

4
🔄 Start Collecting

Tell the program your choice, and it quietly gathers video titles, details, and full spoken words.

5
🧠 Create Smart Search

It builds a clever index that understands the content, ready for your questions.

6
❓ Ask Questions

Type natural questions like 'How to make great coffee?' and see relevant answers pulled from the videos.

🎉 Unlock Insights

Now you have a personal search engine for any YouTube channel, finding exactly what you need instantly.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 20 to 20 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is youtube-rag-scraper?

This Python CLI tool scrapes YouTube channels, playlists, or single videos for metadata like titles, views, and thumbnails, then extracts transcripts using yt-dlp and the YouTube Data API. It chunks transcripts into RAG-ready segments, generates embeddings with sentence-transformers, and builds a FAISS vector index for semantic search over the content. Developers get structured exports in JSON, CSV, or Parquet, plus a simple query command to ask questions like "How does espresso extraction work?" against any channel's library.

Why is it gaining traction?

It handles messy URLs, parallel workers for speed, automatic rate-limit backoff, and resume-from-interrupt, making large-scale YouTube transcript scraping reliable without babysitting. Unlike basic yt-dlp wrappers, the built-in RAG pipeline delivers instant semantic search, saving time on custom vector DB setup. For devs who scrape GitHub repos, profiles, or websites, the one-command knowledge base build is a quick win over manual scripting.

Who should use this?

AI engineers prototyping RAG apps from YouTube channels, like building search over tutorial series. Content analysts scraping YouTube data, transcripts, or thumbnails for insights. Python scripters needing to scrape YouTube videos at scale, similar to scraping GitHub issues or repositories for data pipelines.

Verdict

Grab it for quick YouTube-to-RAG prototypes—solid CLI, docs, and exports make it usable out of the box despite 20 stars and 1.0% credibility score. Too early for heavy production without your own tests, but a strong base to build on.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.