Video Content Analyzer

An integrated workflow for processing video content — extract keyframes, search for related information, store results in Supabase, and publish documentation to Feishu Wiki.

Skills

process_video — Single video analysis

Process a single video file: extract keyframes, search the web for related info, store in Supabase, publish analysis report to Feishu Wiki.

Parameters:

• video_path (string, required): Path to the input video file
• user_id (string, required): UUID of the user running the analysis
• space_id (string, required): Feishu Wiki space ID for publishing

batch_process — Batch directory processing

Process all videos in a directory, auto-classify by filename keywords, store results in Supabase batch_jobs table, and publish categorized analysis reports to Feishu Wiki.

Parameters:

• directory (string, required): Path to directory containing video files
• user_id (string, required): UUID of the user running the analysis
• space_id (string, required): Feishu Wiki space ID for publishing
• categories (object, optional): Mapping of category names to keyword lists for auto-classification
  • Example: {"demo": ["demo", "presentation"], "tutorial": ["tut", "howto"]}
• interval_seconds (integer, optional): Keyframe extraction interval in seconds (default: 10)

Supported video formats: .mp4, .avi, .mov, .mkv, .webm, .flv, .wmv

Batch workflow:

1. Discover all video files in the specified directory
2. Process each video (keyframe extraction → web search → Supabase storage)
3. Auto-classify videos by filename keywords using the categories mapping
4. Publish individual analysis pages to Feishu Wiki
5. Create a category index page per category linking to all its video analyses
6. Track progress in batch_jobs table in Supabase

Environment Variables

CLI Usage

# Single video
python src/main.py single path/to/video.mp4 --user-id \x3Cuuid> --space-id \x3Cspace-id>

# Batch processing
python src/main.py batch /path/to/videos/ --user-id \x3Cuuid> --space-id \x3Cspace-id>

# Batch with categories
python src/main.py batch /path/to/videos/ \
  --user-id \x3Cuuid> \
  --space-id \x3Cspace-id> \
  --categories '{"demo":["demo","presentation"],"tutorial":["tut","howto"]}'