server/skills/video_highlight_pipeline/SKILL.md
Video highlight studio for scene detection, best-frame extraction, scoring, and highlight reel assembly from event/wedding videos
npx skillsauth add proxy2021/enso video_highlight_pipelineInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Video highlight studio for scene detection, best-frame extraction, scoring, and highlight reel assembly from event/wedding videos
Analyze a video file: probe metadata and detect scene change boundaries. Use when: 'analyze this video', 'detect scenes', 'inspect video', 'what's in this video'.
Parameters:
path (string): Path to the video filethreshold (number): Scene detection sensitivity (0.0-1.0, lower = more sensitive, default: 0.3)Extract the N best still frames from the video based on scene boundaries and a scoring strategy (sharpness, face-priority, or balanced).
Parameters:
path (string): Path to the video fileframeCount (number): Number of best frames to extract (default: 20)scoringStrategy (string): Scoring strategy: sharpness, face-priority, or balanced (default: balanced)outputDir (string): Output directory for extracted frames (default: auto)Display extracted frames as a gallery with quality scores and timestamps for review. Allows approve/reject toggling before highlight reel generation.
Parameters:
path (string): Path to the video file (used to load stored frame data)outputDir (string): Directory containing extracted framessortBy (string): Sort frames by: score, timestamp, or rank (default: rank)Assemble a highlight reel video from approved frames/clips with configurable duration, transition style, and audio handling.
Parameters:
path (string): Path to the source video fileduration (number): Target reel duration in seconds (default: 60)clipDuration (number): Duration of each clip segment in seconds (default: 4.5)transitionStyle (string): Transition style: dissolve, fadewhite, smoothleft, mixed (default: mixed)outputDir (string): Output directory for the highlight reelView or update pipeline configuration: scoring weights, frame count, reel duration, transition preferences, and output settings.
Parameters:
action (string): Action: get (view current), set (update settings), reset (restore defaults)frameCount (number): Number of best frames to extract (default: 20)reelDuration (number): Target highlight reel duration in seconds (default: 60)clipDuration (number): Duration per clip in seconds (default: 4.5)transitionDuration (number): Transition duration in seconds (default: 0.5)transitionStyle (string): Transition style: dissolve, fadewhite, smoothleft, mixedscoringStrategy (string): Frame scoring: sharpness, face-priority, balancedsceneThreshold (number): Scene detection threshold (0.0-1.0)testing
Scheduled task health dashboard: monitor task statuses, failure rates, execution history, error classification, circuit breaker states, and drill into specific task run logs. The Team Leader's command center for scheduled task reliability.
testing
Financial accounts dashboard — unified view across brokerage and private-bank accounts with wealth monitoring, refresh logs, and notification settings. Each account is a Cortex entity; each periodic statement is its own synthesis page. Privacy: all data lives at ~/.enso/wiki/ and ~/.enso/data/finances/ (local only — never committed to git).
development
Real-time error monitoring dashboard: error summary with severity breakdown, error trends over time, category analysis, recent error feed, fix tracking, system health score, circuit breaker states, error code analysis, recurring error pattern detection, and actionable recommendations. Reads from Enso's action log, error log, circuit breakers, and error rate monitor.
data-ai
YouTube Manager: subscription management, personalized feed, trending, AI-powered channel discovery, analytics, bulk cleanup