Podcast to Course skill
A skill that turns a podcast transcript or link into framework cards, decision checklists and an interactive HTML course.
- 115 ★ on the repository
- Last commit
Extended access
Why this level
- The skill downloads audio from a link and runs external transcription programs on your machine.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/podcast-to-courseDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The skill downloads audio from a link and runs external transcription programs on your machine.
This is third-party code. Review the repository files before installing.
What it does
The skill turns a podcast into reusable material for a product manager. Input is a transcript or a public episode link (Xiaoyuzhou, a direct `.m4a` or `.mp3` file). The `ingest_podcast.py` script finds the audio, downloads it with resume and transcribes it locally through WhisperKit on Apple Silicon, then aligns speakers and checks quality; fallbacks are Tongyi Tingwu, Feishu Minutes and YouTube transcripts. From the transcript it builds framework cards, decision checklists, interview material and a personal growth log; milestone episodes get an interactive HTML course. The author reports a measurement: a two-hour interview takes about 24 minutes, roughly 5.3 times faster than real time, on one machine. Transcription runs on the device and no paid ASR is needed.
Who it is for. Product managers and teachers who listen to many podcasts and want to keep ready decision tools from them.
Good fit when
- A podcast was heard but a day later the framework cannot be recalled.
- Checklists and interview questions are needed from an interview transcript.
- A Chinese-language episode is transcribed locally without paid ASR.
- A milestone episode is turned into an interactive HTML course.
Not a fit when
- There is no Apple Silicon Mac: local transcription depends on it.
- The podcast is copyright-restricted for processing.
- A verbatim transcript for publication is needed.
Example request
Take the episode at this public link, transcribe it locally with WhisperKit and build framework cards and a decision checklist.Limitations
Local transcription targets Apple Silicon Macs; WhisperKit and SpeakerKit install through `brew` and download models from the network on first run. The 5.3x real-time speed was measured by the author on one machine and yours will differ. Downloading an episode needs a public link. Output rests on the transcript, so recognition errors carry into the cards. No license is set in the repository. Do not process material you have no right to. The last commit dates from August 2026.
How to disable. Delete the `podcast-to-course/` folder from the skills directory and optionally run `brew uninstall whisperkit-cli`. Transcripts and cards sit in your project working folders.
FAQ
What sources does it accept?
A public Xiaoyuzhou link, a direct `.m4a` or `.mp3` link, a ready transcript from any service or your notes.
Is transcription paid?
Not by default: `--provider auto` prefers local WhisperKit and uses paid services only if it is not installed.
What is the output?
A `transcript.md` file, framework cards, decision checklists, interview material, a growth log and optionally an HTML course.
What does local transcription need?
An Apple Silicon Mac, `brew install whisperkit-cli` and the Python dependencies from `requirements.txt`. The first run downloads models.
Does it detect speakers?
Yes, speakers are aligned through SpeakerKit on word timestamps while the words come from WhisperKit.
Discussion
No questions or reviews yet.
Related
Skills14 skills and a CLI save videos, podcasts and articles to a local Markdown base and find quotes in it with a source link
Nine curated skills for content and integrations: Xiaohongshu, video storyboards, Orchard, Chatwoot and copyright filings.
Skill for building study courses: from task and source definition through chapter checks to publishing in Markdown and HTML.
Eleven skills for Claude Code and Codex turn an article or video into Xiaohongshu cards, long images and a narrated short video.
Fourteen skills for media content: articles in five styles, comics, 20 illustration styles, PPT, podcasts and Remotion video.
Twelve Chinese-language skills: HTML pages, long-form writing, coding-agent PRDs, meeting minutes and WeChat publishing.
README in short
The README shows a table of what you provide, what happens and what you get, ways to obtain a transcript and details of the local pipeline with quality checks. Command examples are included.