agentic/code/frameworks/media-curator/skills/assemble/SKILL.md
Assemble media into thematic compilations, concert films, era playlists, and narrative experiences
npx skillsauth add jmagly/aiwg assembleInstall 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.
Assemble collected media into cohesive experiences — narrative compilations, concert films, era playlists, audio documentaries, and thematic collections.
Transform cataloged media into curated experiences that tell stories, preserve performances, or organize creative periods. Outputs range from simple playlists to multi-chapter concert films to narrative documentation.
Content organized in story order (not release order). Useful for concept albums, multi-album sagas, or artists with narrative arcs across their discography.
YAML Structure:
type: narrative
name: "DEMA Saga: Complete Story"
acts:
- name: "Act I: Emergence"
description: "Introduction to the dystopian world"
albums:
- Vessel
- Blurryface
key_tracks:
- "Ode to Sleep"
- "Holding On To You"
- "Car Radio"
- "Stressed Out"
- "Ride"
- "Heavydirtysoul"
- name: "Act II: Escape"
description: "Breaking free from DEMA"
albums:
- Trench
tours:
- "Bandito Tour 2018-2019"
key_tracks:
- "Jumpsuit"
- "Levitate"
- "Morph"
- "My Blood"
- "Chlorine"
- "Pet Cheetah"
- "Bandito"
- name: "Act III: Deception"
description: "Propaganda and false freedom"
albums:
- "Scaled and Icy"
key_tracks:
- "Good Day"
- "Choker"
- "Shy Away"
- "The Outside"
- "Saturday"
- "Mulberry Street"
- name: "Act IV: Resolution"
description: "Final confrontation and liberation"
albums:
- Clancy
key_tracks:
- "Overcompensate"
- "Next Semester"
- "Backslide"
- "Midwest Indigo"
- "Routines in the Night"
- "Vignette"
- "The Craving"
- "Navigating"
- "Snap Back"
- "Oldies Station"
- "At the Risk of Feeling Dumb"
- "Paladin Strait"
output:
playlist: "dema-saga-complete.m3u"
documentation: "dema-saga-narrative-guide.md"
Combine multi-camera footage and audio sources into a cohesive concert experience. Sources may come from multiple shows to create the "perfect" performance.
YAML Structure:
type: concert
name: "The Eras Tour - Opening Night"
date: "2023-03-17"
venue: "State Farm Stadium, Glendale, AZ"
sources:
- camera: "main-stage-center"
path: "footage/main-cam/"
- camera: "b-stage-wide"
path: "footage/b-stage/"
- camera: "audience-left"
path: "footage/aud-left/"
- audio: "soundboard-mix"
path: "audio/sb-mix-2023-03-17.flac"
chapters:
- title: "Lover Era"
start: "00:00:00"
songs:
- "Miss Americana & The Heartbreak Prince"
- "Cruel Summer"
- "The Man"
- "You Need To Calm Down"
- "Lover"
- title: "Fearless Era"
start: "00:23:15"
songs:
- "Fearless"
- "You Belong With Me"
- "Love Story"
- title: "Evermore Era"
start: "00:35:42"
songs:
- "willow"
- "marjorie"
- "champagne problems"
- "tolerate it"
output:
video: "eras-tour-glendale-night1.mp4"
chapters: "eras-tour-glendale-night1.chapters.txt"
playlist: "eras-tour-glendale-night1-songs.m3u"
Best-of from a specific creative period. Combines studio tracks, music videos, live performances, and bonus content.
YAML Structure:
type: era
name: "Reputation Era: Complete Collection"
period: "2017-2018"
components:
studio:
- "reputation (Deluxe Edition)"
videos:
- "Look What You Made Me Do"
- "...Ready For It?"
- "End Game"
- "Delicate"
- "Reputation Stadium Tour (Netflix)"
live:
- "Reputation Stadium Tour - Pasadena Night 1"
- "Reputation Stadium Tour - Pasadena Night 2"
- "Reputation Stadium Tour - Tokyo"
bonus:
- "Making of Reputation"
- "Behind the Scenes: Stadium Tour Rehearsals"
- "Secret Sessions - NYC"
output:
playlist: "reputation-era-complete.m3u"
chapters: "reputation-era-video.chapters.txt"
Narrated experience with music, interviews, rare clips organized into thematic segments.
YAML Structure:
type: documentary
name: "The Rise of Twenty One Pilots"
runtime: "90:00"
segments:
- title: "Origins: The Basement Years"
duration: "15:00"
content:
- narration: "narration/01-origins.mp3"
- music: "No Phun Intended/Drown.mp3"
- music: "No Phun Intended/Save.mp3"
- interview: "interviews/tyler-joseph-early-days.mp3"
- title: "Breakthrough: Regional to Vessel"
duration: "20:00"
content:
- narration: "narration/02-breakthrough.mp3"
- music: "Regional at Best/Holding On To You.mp3"
- music: "Vessel/Car Radio.mp3"
- interview: "interviews/josh-dun-joining-band.mp3"
- title: "World Domination: Blurryface Era"
duration: "25:00"
content:
- narration: "narration/03-blurryface.mp3"
- music: "Blurryface/Stressed Out.mp3"
- music: "Blurryface/Ride.mp3"
- interview: "interviews/grammy-performance.mp3"
output:
audio: "rise-of-top-documentary.mp3"
transcript: "rise-of-top-transcript.md"
Simple M3U playlist generation for any grouping.
YAML Structure:
type: playlist
name: "Taylor Swift - Deep Cuts"
description: "Non-single album tracks that deserve more love"
tracks:
- path: "music/Taylor Swift/Fearless/The Way I Loved You.mp3"
artist: "Taylor Swift"
title: "The Way I Loved You"
album: "Fearless"
- path: "music/Taylor Swift/Red/Sad Beautiful Tragic.mp3"
artist: "Taylor Swift"
title: "Sad Beautiful Tragic"
album: "Red"
- path: "music/Taylor Swift/1989/I Know Places.mp3"
artist: "Taylor Swift"
title: "I Know Places"
album: "1989"
output:
playlist: "taylor-swift-deep-cuts.m3u"
#EXTM3U
#PLAYLIST:DEMA Saga: Complete Story
#EXTART:Twenty One Pilots
#EXTINF:234,Ode to Sleep
/path/to/Vessel/01 Ode to Sleep.mp3
#EXTART:Twenty One Pilots
#EXTINF:197,Holding On To You
/path/to/Vessel/02 Holding On To You.mp3
;FFMETADATA1
title=The Eras Tour - Opening Night
artist=Taylor Swift
date=2023
[CHAPTER]
TIMEBASE=1/1000
START=0
END=1395000
title=Lover Era
[CHAPTER]
TIMEBASE=1/1000
START=1395000
END=2142000
title=Fearless Era
# Generate concat file
cat > concat.txt <<EOF
file 'footage/main-cam/lover-era.mp4'
file 'footage/main-cam/fearless-era.mp4'
file 'footage/main-cam/evermore-era.mp4'
EOF
# Merge with chapters
ffmpeg -f concat -safe 0 -i concat.txt \
-i chapters.txt -map_metadata 1 \
-c copy output.mp4
# Concatenate with 3-second crossfade
ffmpeg -i segment1.mp3 -i segment2.mp3 \
-filter_complex "[0][1]acrossfade=d=3" \
output.mp3
# DEMA Saga: Complete Narrative Guide
## Overview
The DEMA Saga spans four studio albums and tells the story of a dystopian city-state called DEMA, ruled by nine bishops who control citizens through propaganda and fear. The protagonist's journey from captivity to liberation mirrors themes of mental health, artistic expression, and finding authentic community.
## Key Characters
- **Tyler Joseph (Clancy)**: The narrator and protagonist
- **Josh Dun**: The Torchbearer, guide to freedom
- **The Nine Bishops**: Led by Nico and the Niners, represent control and conformity
- **The Banditos**: Rebels who escaped DEMA
## Album-by-Album Guide
### Act I: Emergence (Vessel, Blurryface)
**Vessel (2013)** introduces the internal struggle...
**Blurryface (2015)** personifies anxiety and depression as a character...
### Act II: Escape (Trench)
**Trench (2018)** reveals the geography of DEMA...
[Continue with detailed analysis]
aiwg assemble --type playlist \
--name "Best of 2023" \
--collection ~/Music/2023 \
--output best-of-2023.m3u
aiwg assemble --type concert \
--name "Eras Tour - Night 1" \
--collection ~/Videos/ErasTour/2023-03-17 \
--output eras-night1.mp4
aiwg assemble --type narrative \
--name "DEMA Saga" \
--collection ~/Music/TwentyOnePilots \
--output dema-saga/
All commands generate:
Files are written to --output path or default location based on type.
data-ai
Report which research-corpus radar sidecars are overdue for refresh. Computes staleness (days since last refresh vs the cadence window) for every radar, sorted most-overdue-first. Runs via `aiwg corpus radar-status`.
data-ai
Aggregate research-corpus radar sidecars into a corpus or per-cluster freshness report — totals, overdue count, per-cluster / per-GRADE / per-trajectory breakdowns, an overdue table, and per-radar rationale snippets. Runs via `aiwg corpus radar-report`.
testing
Scaffold radar/freshness sidecars for research-corpus REFs. Pulls title/authors from the citation sidecar and GRADE from the analysis doc, defaults the refresh cadence from GRADE and the cluster from a corpus-local map, and stamps documentation/radar/REF-XXX-radar.md. Runs via `aiwg corpus radar-init`.
data-ai
Compute an entity's publication trajectory — per-year paper counts, topic drift, hot-streak detection (≥3 consecutive A-grade years), and career phase. Runs via `aiwg corpus profile-temporal`.