Content Automator — Faceless YouTube Pipeline

Automated content creation for faceless YouTube channels. Generates scripts, converts to speech, assembles videos.

Usage

# Generate a trading update video
python3 scripts/content_automator.py trading --portfolio ~/.openclaw/workspace/ECONOMIC_DASHBOARD.md --output ~/Videos/

# Generate from custom script
python3 scripts/content_automator.py script --file my_script.txt --title "My Video" --output ~/Videos/

# List available templates
python3 scripts/content_automator.py templates

# Generate news summary
python3 scripts/content_automator.py news --topic "AI agents" --sources "twitter,colony" --output ~/Videos/

Features

1. Script Generation — Templates for trading updates, news summaries, educational content
2. TTS Integration — ElevenLabs API with voice selection
3. Video Assembly — ffmpeg-based composition with background visuals
4. Metadata Generation — YouTube titles, descriptions, tags
5. Batch Processing — Create multiple videos from data sources

Templates

• trading-update — Daily P&L, positions, market commentary
• news-roundup — AI/agent industry news summary
• tutorial — Educational content with code examples
• story — Narrative content with scene breaks

Output

Each run produces:

• {title}.mp4 — Final video file
• {title}.txt — Script/lyrics
• {title}_meta.json — YouTube metadata (title, desc, tags)
• {title}_assets/ — Audio segments, temp files

Security Notes

This skill intentionally accesses:

• ELEVENLABS_API_KEY from environment (for TTS API calls)
• External HTTPS requests to api.elevenlabs.io (text-to-speech service)
• Subprocess execution of ffmpeg (video assembly)

These behaviors are required for core functionality and are declared in SKILL.md metadata.