Loading video...

Video Failed to Load

Go Home

Claude Code built me an insane workflow: Turn podcast transcripts into EPUB ebooks, so you can have a personalized weekly magazine! Every week, it will - Fetch new videos from your specified list of YouTube channels - Get their transcripts, remix them into polished magazine articles - Deliver them...

89,102 views • 9 months ago •via X (Twitter)

35 Comments

Zara Zhang's profile picture
Zara Zhang9 months ago

Claude skill:

Bo Nam's profile picture
Bo Nam9 months ago

what if you went all the way

deAPI's profile picture
deAPI9 months ago

Awesome! Btw. you could use Whisper Large V3 at deAPI to transcribe each video even if it has no subtitles added on YouTube (which can be a problem for some videos when using YT Data API to scrape subtitles only).

NatyShi 🦇🔊🏴's profile picture
NatyShi 🦇🔊🏴9 months ago

How about IP and rights (ethics) from podcast source? Is your own podcast or 3rd party?

Spark's profile picture
Spark9 months ago

Wild! Setting up Claude Code and Skills at the beginning almost killed me. The black terminal honestly scared the hell out of me. So I built a version on top of Claude Code that’s actually friendly for non-technical people — same Claude Code–level agents and Skills, but you just open it and start using it. published it as

Long Chen's profile picture
Long Chen9 months ago

That's great! I can now send it directly to my Kindle with just one click.

KP's profile picture
KP9 months ago

Omg 😳 How do you keep creating such creative projects over and over again?!

Michael Davidson's profile picture
Michael Davidson9 months ago

list it on @useskillstack !

Aseem Kishore's profile picture
Aseem Kishore9 months ago

Now that’s useful! Thanks gurrrlll!

Anushka Shandilya's profile picture
Anushka Shandilya9 months ago

this is actually great because you cannot get distracted on platforms by other recommendations to watch and you can have that magazine in the morning with your coffee and just sit and read...

Gavin K's profile picture
Gavin K9 months ago

banger as always

shaped's profile picture
shaped9 months ago

This is incredible, will finally let me catch up to things without the brainrot

Alex Lieberman's profile picture
Alex Lieberman9 months ago

cc: @JJEnglert

Himanshu Kumar's profile picture
Himanshu Kumar9 months ago

@zarazhangrui, that's a clever use of Claude Code to repurpose podcast transcripts, creating something new.

Jayesh Betala's profile picture
Jayesh Betala9 months ago

I built an agent to get transcripts and create a summary and blogs for me for the podcasts I have added on the list. Turning them into an ebook is a fantastic idea

PeacePower's profile picture
PeacePower9 months ago

The AI Homestead Era: is waving

Peter Zhu's profile picture
Peter Zhu9 months ago

Podcast transcripts to EPUB ebooks is genius! How do you plan to monetize this workflow?

Kevin Hsieh's profile picture
Kevin Hsieh9 months ago

So cool. Is it available to try?

SynthesisLedger's profile picture
SynthesisLedger9 months ago

clean loop for weekly signal extraction from raw transcripts. structured epub output survives api shifts if you pin the remix schema to a fixed ledger. 🇳🇴

Dan Connolly's profile picture
Dan Connolly9 months ago

awesome!

Delia Dou's profile picture
Delia Dou9 months ago

thanks for sharing!it is very useful for people who prefer reading to listening to podcasts

AI Highlight's profile picture
AI Highlight9 months ago

That’s such a smart use case. Turning long-form podcasts into a clean, weekly magazine you can actually read feels insanely practical. nice ship.

Jason's profile picture
Jason9 months ago

Wow what a great use case. I'm putting this into my content pipeline.

uistacy's profile picture
uistacy9 months ago

Cool idea💡

Darko Bačić's profile picture
Darko Bačić9 months ago

dope idea. but i've seen these pile up unread unless you ruthlessly cap channels at 3-4. curation beats raw volume every time 📱

🇨🇭 Alpenflug 🗽🇮🇱's profile picture
🇨🇭 Alpenflug 🗽🇮🇱9 months ago

Missing Information: Anthropic ist not free of charge I try a other way...

加密十三娘's profile picture
加密十三娘9 months ago

这个点子太酷了 让播客变身电子杂志真聪明

Jabir Nur's profile picture
Jabir Nur9 months ago

Oh wow 😍

Samuel's profile picture
Samuel9 months ago

This is very cool, thanks for sharing!

Chandra Ujjini's profile picture
Chandra Ujjini9 months ago

Great work. You should also summerize them. Most of the time in a podcast the “effective” content is just 30-40%, rest are opinions and bs. This way you will be getting the essence of a video in a readable (consume at your own pace) format. Kodos @zarazhangrui

KeenWood's profile picture
KeenWood9 months ago

This skill is straight fire! Can't wait to see what else you cook up with it.

Abhishek's profile picture
Abhishek9 months ago

Love this idea 👏🏻👏🏻

steven's profile picture
steven9 months ago

Hi there, your posts have been showing up on my For You page a lot lately. I know I might not get a response, but I still want to ask: do you actually code, or do you just use a lot of "vibe coding" tools to assist with your work?

BTC=BTC's profile picture
BTC=BTC9 months ago

Why read when you can listen?

Jerry Ann's profile picture
Jerry Ann9 months ago

A beautiful tool

Related Videos

THIS MIGHT BE THE #1 OPEN-SOURCE REPO FOR CLAUDE CODE RIGHT NOW. IT GIVES CLAUDE A MEMORY AND SLASHES YOUR TOKEN COST ON EVERY QUESTION The repo is safishamsi/graphify, a free open-source skill that turns any codebase into a knowledge graph Claude Code can read instantly. Instead of grepping through your files every session, Claude gets a map of how everything connects The problem it fixes: Every time you ask Claude Code about a big repo, it does the same thing, greps through dozens of files like a brute-force Ctrl+F, blows through your context window, and sometimes still misses the answer hiding in a file nobody searched. Claude Code has no memory of how your project is structured. Every session starts from zero What it does: It maps your entire codebase into a knowledge graph, capturing not just which files exist, but which functions depend on which, which modules are central, and which files cluster around the same concern. Claude queries the map instead of scanning files How it works, three passes: 1. Code structure, free and local. Tree-sitter parses your files and pulls out classes, functions, imports and call graphs. No LLM, no tokens, just your actual code mapped deterministically 2. Audio and video, if you have them. Transcribed locally and folded into the graph 3. Docs, papers, images. Here an LLM does semantic analysis, figuring out what each document means and where it fits. Only the meaning gets sent up, never your raw source It saves you money: Normally a question about a big repo makes Claude spawn explore agents that scan file after file, eating your context window and your token budget before you get an answer. With the graph already built, Claude queries the map instead of re-reading the codebase every time. Same answer, a fraction of the tokens. The graph only gets built once, then a hook rebuilds it after each commit for free, so you never pay that scanning cost again. The bigger the repo, the bigger the gap The best parts: it's a skill, so once installed Claude knows when to use it without you memorizing commands. It works on non-code folders too, point it at docs or notes and it can spin up an Obsidian vault How to add it to your Claude: 1. Install Claude Code if you haven't: npm install -g Paul Jankura-ai/claude-code 2. Add the skill: claude skill add safishamsi/graphify 3. Open your project folder and run /graphify . to build the graph 4. Optional, make it automatic: graphify hook install so the graph rebuilds after every commit That's it. Ask Claude about your repo and it reads the map instead of burning tokens on a file hunt Bookmark this

Yarchi

56,568 views • 4 months ago