ワンクリックで
start
Set up the project and learn how to use it. Run this first when opening the project.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Set up the project and learn how to use it. Run this first when opening the project.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Manage TikTok profiles to scrape. Switch accounts, add multiple profiles, or list saved profiles.
Transcribe ALL videos from a TikTok profile. Use when you want to process an entire profile's content.
Transcribe specific video URL(s). Use when you have one or more TikTok video URLs to process.
Turn a transcript summary into a reusable Claude skill. Researches the topic and creates a skill file.
| name | start |
| description | Set up the project and learn how to use it. Run this first when opening the project. |
| user_invocable | true |
When the user runs /start, follow this workflow:
Display this welcome message:
Welcome to the Short-Form Video Transcriber!
This project helps you:
Let me check if your environment is set up...
Run these checks:
# Check Python
python3 --version
# Check ffmpeg
which ffmpeg || where ffmpeg
# Check if venv exists
ls -la .venv/bin/activate 2>/dev/null || echo "NO_VENV"
Create and set up the environment (this also installs yt-dlp automatically):
python3 -m venv .venv
source .venv/bin/activate # On Windows: .venv\Scripts\activate
pip install -e ".[dev]"
pip install yt-dlp # Install yt-dlp in the venv
Activate and ensure yt-dlp is installed:
source .venv/bin/activate # On Windows: .venv\Scripts\activate
pip install yt-dlp # Ensure yt-dlp is available
pytest tests/unit/ -v --tb=short 2>&1 | tail -5
ffmpeg is required for audio extraction. Provide platform-specific instructions:
macOS:
brew install ffmpeg
Linux (Debian/Ubuntu):
sudo apt update && sudo apt install ffmpeg
Linux (Fedora/RHEL):
sudo dnf install ffmpeg
Windows:
winget install ffmpeg
Or download from https://ffmpeg.org/download.html and add to PATH.
After installing ffmpeg, re-run /start to continue setup.
After setup is verified, explain the commands:
Setup complete! Here's how to use this project:
| Command | What it does |
|---|---|
/start | You're here! Set up the project and see available commands |
/bulk | Transcribe ALL videos from a TikTok profile |
/transcribe | Transcribe specific video URL(s) you paste |
/accounts | Switch profiles, add multiple accounts, process several at once |
/skillify | Turn a summary into a reusable Claude skill |
/bulk - Process Entire ProfileUse this when you want to transcribe all videos from someone's TikTok.
Just run /bulk and I'll ask you for:
https://www.tiktok.com/@example_creator)I'll then download, transcribe, and summarize each video automatically.
/transcribe - Process Specific VideosUse this when you have specific video URLs you want to transcribe.
Just run /transcribe and paste the URLs:
https://www.tiktok.com/@username/video/123456789
https://www.tiktok.com/@username/video/987654321
You can paste one URL or multiple URLs at once.
/accounts - Manage Multiple ProfilesWant to scrape from different TikTok accounts? Use /accounts to:
Your saved profiles are stored in accounts.json.
/skillify - Create Claude Skills from SummariesTurn any summary into a reusable Claude skill! This command:
Great for building a knowledge base from video content!
All commands create:
transcripts/ - Raw transcripts with metadatasummaries/ - Organized summaries grouped by topicsummaries/INDEX.md - Master index of all content/skillify go to your chosen directoryWhat would you like to do?
/bulk to process an entire TikTok profile/transcribe to process specific video URLs/skillify to turn a summary into a Claude skillAfter explaining, wait for the user to either:
/bulk or /transcribe/transcribe)If they paste URLs directly without a command, process them as if they ran /transcribe.
If setup fails:
If tests fail:
pip install -e ".[dev]"