- Home
- Skills
- Social Media
- YouTube Knowledge Factory
More screenshots
Works with the AI Agents you already use
YouTube Knowledge Factory
Capture subtitles from ALL the videos on your favorite channels and convert them into clean text for RAG—without watching a single video.
$7.99
YouTube Knowledge Factory
Example session with this skill installed
"Search my archive for what the ES devs said about Rust async"
Result: 3 results with links to videos, % match, title, channel, date
- Read your context and instructions
- Compiled the youtube knowledge factory
I have initialized the radar for the specified channel. The system is currently capturing subtitles, converting them to clean Markdown, and indexing them for the yt-ask tool.
Capture Status & Research Findings
| Channel | Status | Videos Processed | RAG Ready |
|---|---|---|---|
| Lex Fridman | Processing | 50 / 50 | Yes |
| Total Items | Complete | 50 | 100% |
Top Search Matches for "Artificial Intelligence"
| Video ID | Title Extract | Key Insights |
|---|---|---|
vid_001 | Sam Altman: OpenAI and the Future | Discussion on AGI timelines and compute scaling laws. |
vid_012 | Yann LeCun: Dark Matter of Intelligence | Argument for world models over current LLM architectures. |
vid_045 | Demis Hassabis: DeepMind | Focus on protein folding and AI for scientific discovery. |
# Example generated Markdown Frontmatter (srt2md.py output)
title: "Sam Altman: OpenAI and the Future of AI"
channel_id: "UCv0n78qqWp29G-S"
date_captured: 2023-10-27
tags: [ai, openai, agi]
---
[Cleaned transcript text continues here...]
Next steps
- Run
yt-ask "What is the consensus on AGI?"to query the new data. - Access the live dashboard at
localhost:8002to browse the full transcripts. - Add more channels to
lib/canales.shto expand your research factory.
Connects securely to your tools. The creator never sees your data.
About this skill
Build your own hemeroteca of YouTube channels. Capture subtitles from ALL videos of your favorite channels and convert them to clean markdown for RAG. Without watching a single video.
-
Radar secuencial: 1 channel → 1 video with sleeps, lock, blocklist with cause, breaker x3, IPv4 anti-ban
-
srt2md: srt → clean md with YAML frontmatter (no timestamps, dedup)
-
Dashboard :8002 live (progress per channel, explorer, search)
-
yt-ask: ask your hemeroteca like local RAG ("What did X say about Y?")
-
Cron daily + automatic backup
-
You follow 10+ YouTube channels and don't have time to watch
-
You want to research without losing time
-
You need a personal knowledge base from channels you watch
-
You want to capture entire channels automatically (anti-ban included)
This skill does NOT include any video transcripts or subtitles. You build your own hemeroteca with channels you follow. The 3 demo channels in lib/canales.sh are for testing only — replace them with your own channels.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Social Media.
Cannot download private, members-only, or Shorts videos. Requires manual entry of channel IDs in config. User is responsible for YouTube ToS compliance.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 15 days ago
- Accounts
GitHub
- Software
Python
Permissions
Allowed Hosts