nuva-lab
16 verified skills80 total stars
extract-clip
Extract a video segment using FFmpeg with precise start/end times
tools5
analyze-video
Analyze raw video content using Gemini to identify speakers, topics, key moments, and potential clip opportunities
tools5
skills/remotion-render
# Remotion Render Skill Generic video rendering with motion graphics overlays. ## Components - `VideoClip.tsx` - Main composition (video + captions + annotations) - `components/RollingCaption.tsx` - Karaoke-style word highlighting - `components/SpeakerLabel.tsx` - Positioned speaker annotations - `components/ContextBadge.tsx` - Location/event badge ## Usage Called by `make-video` skill. Not typically invoked directly. ```bash npx remotion render VideoClip output.mp4 --props=props.json ```
tools5
write-script
Generate voiceover scripts in Joyce's style for video clips
tools5
separate-audio
Text-guided audio source separation using SAM-Audio via mlx-audio
documentation5
skills/talking-head
# Talking-Head Video Processing Process direct-to-camera videos with sentence-level precision, rolling captions, and multi-format output. ## Full Pipeline Overview ``` Raw Video (47+ min) ↓ smart_chunk.py Chunks (2.5-3.5 min) ↓ mlx_transcribe.py --word-timestamps Transcript (word-level timestamps) ↓ sentence_split.py (precise cuts, no overlap) Sentence Clips (~10s each, frame-accurate) ↓ analyze_script.py (RECALL-FIRST) Topics with complete arcs (hook → elaborate → conclude)
tools5
skills/validate-media
# Validate Media Skill Pre-flight media validation and diagnostics using ffprobe. ## Purpose Check video/audio files for common issues before rendering: - Duration mismatches between video and audio tracks - Missing audio tracks - Codec compatibility - Volume levels - Potential freeze points ## Usage ```bash python skills/validate-media/validate.py <video_file> [--verbose] ``` ## Output JSON report with issues and recommendations: ```json { "file": "video.mp4", "video_duration": 35.1
development5
audio-process
Audio processing utilities - noise reduction, normalization, enhancement
content-media5
skills/make-video
# Make Video Skill Single-script video production from project config. ## Usage ```bash python skills/make-video/make_video.py video_projects/space_investing/ ``` That's it. One command, one output. ## Project Structure ``` video_projects/<name>/ ├── project.json # Config (required) ├── source_video.mp4 # Input video (required) ├── voiceover.wav # Audio (optional, uses video audio if missing) └── output/ └── final.mp4 # Generated output ``` ## project.json ```json { "
development5
transcribe-audio
ASR with ~30ms timestamp precision using Qwen3-ASR + ForcedAligner
testing5
voice-clone
Clone a voice using qwen3-tts and generate speech from text
tools5
align-captions
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'.
content-media5
transcribe-clip
Transcribe a video clip using Gemini to get timestamped segments for captions
tools5
find-golden-segments
Find naturally clean, coherent video segments worth keeping (selection over repair)
development5
skills/inspect-video
# Inspect Video Skill Gemini-powered video inspection for quality analysis and debugging. ## Purpose Use Gemini 3 Pro's video understanding to: - Describe what's happening in the video - Identify freeze frames, black frames, glitches - Detect visual quality issues - Verify content matches expectations - Generate quality reports ## Usage ```bash python skills/inspect-video/inspect.py <video_file> [--prompt "custom query"] ``` ## Default Inspection Without custom prompt, runs a standard qua
development5
skills/chunk-process
# Chunk Process Skill Smart video chunking and MLX-accelerated transcription for long-form content. ## Problem Solved - Raw footage too long for single Gemini upload (~47 min = 5GB+) - Need word-level timestamps for precise cutting - Fixed-length chunks break mid-sentence ## Smart Chunking Instead of fixed 5-minute segments, `smart_chunk.py` finds natural break points: ```bash python skills/chunk-process/smart_chunk.py raw_footage.mp4 -o chunks/ ``` **How it works:** 1. Detect silence regi
development5