bulk

name	bulk
description	Transcribe ALL videos from a TikTok profile. Use when you want to process an entire profile's content.
user_invocable	true

Bulk Video Transcription

This command processes ALL videos from a TikTok profile - scraping URLs, downloading audio, transcribing, and creating organized summaries.

Workflow

Step 1: Get Profile URL

Ask the user:

What TikTok profile do you want to process?

Example: https://www.tiktok.com/@example_creator

Step 2: Get Limit (Optional)

Ask the user:

How many videos do you want to process?

Options:

• All - Process every video on the profile
• Number - Process only the first N videos (e.g., "5" or "10")

Step 3: Verify Environment

Check that the project is set up:

cd /path/to/project
source .venv/bin/activate
which yt-dlp

If not set up, tell user to run /start first.

Step 4: Scrape Video URLs

Run the scraper to discover all videos:

source .venv/bin/activate && python -c "
from short_form_scraper.scraper.tiktok import TikTokScraper

scraper = TikTokScraper('PROFILE_URL_HERE')
videos = list(scraper.get_video_urls(limit=LIMIT_OR_NONE))

print(f'Found {len(videos)} videos:')
for i, v in enumerate(videos, 1):
    print(f'{i}. {v.title[:60]}...' if len(v.title) > 60 else f'{i}. {v.title}')
    print(f'   URL: {v.url}')
    print()
"

Show the user the list and confirm they want to proceed.

Step 5: Download and Transcribe Each Video

For each video, run:

source .venv/bin/activate && python -c "
from short_form_scraper.scraper.tiktok import TikTokScraper
from short_form_scraper.downloader.video import VideoDownloader
from short_form_scraper.transcriber.whisper import WhisperTranscriber
from pathlib import Path
import json

# Initialize components
scraper = TikTokScraper('PROFILE_URL_HERE')
downloader = VideoDownloader()
transcriber = WhisperTranscriber()

# Create output directories
Path('transcripts').mkdir(exist_ok=True)
Path('state').mkdir(exist_ok=True)

# Get videos
videos = list(scraper.get_video_urls(limit=LIMIT_OR_NONE))

for i, metadata in enumerate(videos, 1):
    print(f'[{i}/{len(videos)}] Processing: {metadata.title[:50]}...')

    try:
        # Download audio
        audio_path = downloader.download(metadata.url, Path(f'state/audio_{metadata.id}'))
        print(f'  Downloaded: {audio_path}')

        # Transcribe
        transcript = transcriber.transcribe(audio_path)
        print(f'  Transcribed: {len(transcript)} chars')

        # Save transcript with metadata
        transcript_file = Path('transcripts') / f'{metadata.id}.txt'
        content = f'''Video ID: {metadata.id}
Title: {metadata.title}
URL: {metadata.url}
Duration: {metadata.duration}s

--- TRANSCRIPT ---

{transcript}
'''
        transcript_file.write_text(content)
        print(f'  Saved: {transcript_file}')

        # Clean up audio file to save space
        audio_path.unlink()

    except Exception as e:
        print(f'  ERROR: {e}')

    print()

print('Done! Transcripts saved to transcripts/')
"

Step 6: Summarize Transcripts (Claude Code)

After all transcripts are generated, YOU (Claude Code) will:

1. Read each transcript file from transcripts/ directory

2. For each transcript, analyze and extract:

   • Topic: A descriptive 2-4 word kebab-case topic name (e.g., "agentic-engineering", "context-management")
   • Summary: One-sentence summary of the main point
   • Key Tips: 3-5 actionable bullet points
   • Details: Additional context

3. Create organized output in summaries/ directory:

   summaries/
   ├── INDEX.md
   ├── {topic-name}/
   │   └── {slugified-title}.md
   

   IMPORTANT: Filename from Video Title

   • Extract the video title from the yt-dlp metadata (stored in the transcript file header)
   • Convert to kebab-case slug: lowercase, spaces to dashes, remove special chars
   • Example: "How to Use Context Windows" → how-to-use-context-windows.md
   • Example: "Claude Code Tips & Tricks!" → claude-code-tips-tricks.md

4. Each summary file format:

   ---
   video_id: {id}
   title: {title}
   url: {url}
   topic: {topic}
   ---
   
   # {Title}
   
   ## Summary
   {one-sentence summary}
   
   ## Key Tips
   - {tip 1}
   - {tip 2}
   - {tip 3}
   
   ## Details
   {additional context}
   
   ## Full Transcript
   {original transcript}
   

5. Create INDEX.md listing all summaries grouped by topic:

   # Video Summaries Index
   
   ## {Topic Name}
   - [{Video Title}](./{topic-name}/{slugified-title}.md)
   

Step 7: Report Results

After completion, report:

• Total videos found
• Successfully transcribed
• Failed (if any)
• Topics identified
• Output locations

Error Handling

• If a video fails, log the error and continue with the next one
• Save partial progress (transcripts saved immediately)
• User can re-run and existing transcripts won't be re-downloaded