# AI use cases for Podcast Shorts Factory (krakonjac300-pixel/podcast-shorts-factory)

Repo: [krakonjac300-pixel/podcast-shorts-factory](https://github.com/krakonjac300-pixel/podcast-shorts-factory) | Use cases: 3 | Page: https://dedupedai.com/use-cases/krakonjac300-pixel-podcast-shorts-factory/

Podcast Shorts Factory is a Python pipeline of ten cooperating AI agents that turns a YouTube podcast link into captioned vertical shorts and can schedule them to YouTube Shorts, configured for free AI providers.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Turn a podcast episode into vertical shorts in one command: https://dedupedai.com/use-cases/krakonjac300-pixel-podcast-shorts-factory/podcast-to-shorts-in-one-command/
- Approve clip candidates yourself before rendering: https://dedupedai.com/use-cases/krakonjac300-pixel-podcast-shorts-factory/review-clip-candidates-before-rendering/
- Run a fully unattended daily shorts schedule: https://dedupedai.com/use-cases/krakonjac300-pixel-podcast-shorts-factory/unattended-daily-shorts-schedule/

## How to turn a podcast episode into vertical shorts in one command with Podcast Shorts Factory

Repo: [krakonjac300-pixel/podcast-shorts-factory](https://github.com/krakonjac300-pixel/podcast-shorts-factory) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README

Give it a YouTube podcast link and get transcribed, scored, cut, captioned 9:16 clips.

**Who it is for:** A podcaster who publishes weekly and never has time to cut clips.

**The problem:** Making shorts by hand means scrubbing hours of audio for good moments.

**The outcome:** Finished vertical clips with karaoke captions, music and titles ready to post.

### Steps

1. On Windows install Python and ffmpeg with `winget install -e --id Python.Python.3.11` and `winget install -e --id Gyan.FFmpeg`
2. Run the setup wizard to pick a provider, key, niche and platforms
3. Run `./factory.sh auto "https://www.youtube.com/watch?v=VIDEO_ID"` (or `.\factory auto` on Windows)
4. Check the rendered clips in the output folder

### Prompt to paste into your agent

```text
Run krakonjac300-pixel/podcast-shorts-factory with `./factory.sh auto "<YouTube URL>"`. Report how many clips passed QA and confirm each is vertical 9:16 with captions.
```

### Caveat

It ships configured for free AI providers, whose rate limits can slow long episodes.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/krakonjac300-pixel-podcast-shorts-factory/podcast-to-shorts-in-one-command/

## How to approve clip candidates yourself before rendering with Podcast Shorts Factory

Repo: [krakonjac300-pixel/podcast-shorts-factory](https://github.com/krakonjac300-pixel/podcast-shorts-factory) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README

Run the find step, approve or reject the scored moments, then render only the ones you like.

**Who it is for:** A show host who wants control over which moments go public.

**The problem:** Fully automatic picks can surface moments you would never post.

**The outcome:** Only approved moments get rendered and finished.

### Steps

1. Run the find command to download, transcribe and score candidates
2. Run the review command to approve or reject each candidate
3. Run edit to render approved clips, then finish for the QA review
4. Run upload to post or export

### Prompt to paste into your agent

```text
Using podcast-shorts-factory, run find on <YouTube URL>, list the scored candidates with timestamps, and wait for my approvals before running edit. Confirm only approved clips were rendered.
```

### Caveat

Each stage is a separate command, so a skipped finish step blocks posting.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/krakonjac300-pixel-podcast-shorts-factory/review-clip-candidates-before-rendering/

## How to run a fully unattended daily shorts schedule with Podcast Shorts Factory

Repo: [krakonjac300-pixel/podcast-shorts-factory](https://github.com/krakonjac300-pixel/podcast-shorts-factory) | Difficulty: Advanced | Time: 1 hour 30 minutes | Basis: From the README

Use the daily command to pull trends, grab the newest episode, cut clips and schedule posts to YouTube Shorts.

**Who it is for:** A clip channel operator who wants posts going out every day without touching a PC.

**The problem:** Daily posting by hand is a full-time job.

**The outcome:** Shorts scheduled server-side on YouTube, plus TikTok and Reels exports.

### Steps

1. Complete setup including YouTube upload access
2. Run the daily command once and confirm the scheduled posts
3. Run stats to refresh metrics so the Manager updates its learnings
4. Schedule the daily command with your OS scheduler

### Prompt to paste into your agent

```text
Configure podcast-shorts-factory for <channel>, run the daily command, and list the posts it scheduled with their times. Verify they appear as scheduled in YouTube Studio.
```

### Caveat

Only clip episodes you own or have rights to repost; the tool will download any YouTube link you give it.

Risk, credentials: Uploading uses your YouTube account through the official API, so the stored token can post to your channel.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/krakonjac300-pixel-podcast-shorts-factory/unattended-daily-shorts-schedule/