Menu
Generate voiceover scripts in Joyce's style for video clips
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Generate karaoke-style word-level timestamps by aligning script text to audio using Qwen3-ForcedAligner + jieba for Chinese word segmentation. Use when the user says 'align captions', 'karaoke timestamps', 'word timestamps', 'caption alignment', 'sync text to audio'.
Analyze raw video content using Gemini to identify speakers, topics, key moments, and potential clip opportunities
Audio processing utilities - noise reduction, normalization, enhancement
Extract a video segment using FFmpeg with precise start/end times
Find naturally clean, coherent video segments worth keeping (selection over repair)
Text-guided audio source separation using SAM-Audio via mlx-audio
Based on SOC occupation classification
| name | write-script |
| description | Generate voiceover scripts in Joyce's style for video clips |
Use this skill to create voiceover scripts for video clips, matching Joyce's energetic insider style.
# Generate script prompt from golden segments
python skills/write-script/write_script.py golden_segments.json
# With custom angle/thesis
python skills/write-script/write_script.py golden_segments.json --angle "space investment"
write_script.py with golden segments JSONHOOK (5-10s) → Grab attention, conflict/contrast
CONTEXT (10-15s) → Who/what/where, establish scene
INSIGHT (15-30s) → Key quote/moment, why it matters
ANALYSIS (10-15s)→ What this means, your take
PIVOT (5-10s) → Connect to broader trend
See style_guide.md for Joyce's full style profile including: