Skip to content

Home

Jump to bottom
Jaydeep Solanki edited this page Mar 8, 2026 · 9 revisions

yt-study Wiki

Practical documentation for using, configuring, extending, and maintaining yt-study.

yt-study turns YouTube videos, playlists, and URL batches into Markdown study notes using transcript extraction, LLM generation, and a Rich terminal dashboard.

  • Typer for the CLI
  • Rich for the live terminal dashboard
  • LiteLLM for model/provider abstraction
  • youtube-transcript-api for transcript retrieval
  • pytubefix for metadata, playlist expansion, and chapter discovery

Start Here

Page Best for What you get
Installation New users Install methods, first run, config path, Windows notes
Usage Everyday use Commands, flags, examples, output behavior
Configuration Power users Real config keys, precedence, provider mapping
Troubleshooting Failed runs Recovery steps for API, transcript, playlist, and terminal issues
FAQ Quick answers Short answers to common questions
Architecture Contributors Internal flow, module boundaries, event model
Development Maintainers Make targets, hooks, tests, docs workflow

Project Snapshot

Area Details
Inputs Single video URL, playlist URL, batch file
Outputs One Markdown file per video, or per-chapter Markdown for long videos
UI Typer CLI + Rich live dashboard
LLM Layer LiteLLM
YouTube Data youtube-transcript-api + pytubefix
Runtime Config ~/.yt-study/config.env
Logging ~/.yt-study/logs/yt-study-{timestamp}.log

Common Paths

Runtime config: ~/.yt-study/config.env
Logs:           ~/.yt-study/logs/
Wiki source:    wiki/  (Git submodule)
Main package:   src/yt_study/
Tests:          tests/

Main User Flows

First-time setup

  1. Install the package.
  2. Run yt-study setup.
  3. Configure model, API key, output directory, and concurrency.
  4. Run yt-study process "URL".

Single video

  1. Parse the URL.
  2. Fetch title, duration, chapters, and transcript.
  3. Choose chapter mode or chunked mode.
  4. Write Markdown output.

Playlist or batch file

  1. Expand the input into multiple video jobs.
  2. Process videos concurrently through the core pipeline.
  3. Show progress in the dashboard.
  4. Print a success/failure summary.

For Contributors

Notes

  • The wiki is maintained as a Git submodule.
  • Root docs and wiki pages should stay aligned on commands, config keys, and output behavior.
  • If a feature changes, update both the concise docs surface and the deeper reference pages.

Clone this wiki locally