I was in the process of creating an app that would create short social media videos that I can post to various social media platforms like Twitter/X, Facebook, Instagram, and others. I'll share the description of the subagents and components of it right here:
OVERVIEW OF SUBAGENTS: {no overview was provided in the original text}
SUBAGENT 1: "Short Script Creator"
• Skill: #190 - Write or rewrite text based on instructions
• Purpose: Generates the short promotional script (hook, CTA, hashtags), including rough timing (e.g. "Hook at 0s–4s," "CTA at 5s–9s," "Hashtags at 10s–14s").
• Input: Topic/keywords and any user instructions.
• Output: Brief text (hook, CTA, hashtags) with approximate timing.
SUBAGENT FINAL OUTPUT: [short-promotional-script]
SUBAGENT 2: "B-roll Finder & Editor"
• Skills:
1) #205 - Get Royalty-Free VERTICAL Videos For A Given Keyword (to find vertical footage)
2) #194 - Cut Small Section From MP4 Video (to trim one chosen clip to 10–15s)
• Purpose: Locates relevant vertical footage and trims it to fit the desired length.
• Input: Keywords from the script; desired duration (10–15s).
• Output: One trimmed MP4 clip of vertical b-roll.
SUBAGENT FINAL OUTPUT: [trimmed-vertical-broll]
SUBAGENT 3: "Music Curator"
• Skill: #220 - Generate 3x WAV Sound Effects From Text Prompt
• Purpose: Produces short background music/sound effect (7–15s) matching the intended mood.
• Input: Brief description of the desired style or tone ("dramatic," "upbeat," etc.).
• Output: Three .wav files (each 7–15s). One is chosen for final use.
SUBAGENT FINAL OUTPUT: [short-bgm-wav-files]
SUBAGENT 4: "Overlay Instruction Generator"
• Skill: #190 - Write or rewrite text based on instructions
• Purpose: Converts the short script into a simple timed transcription or overlay plan suitable for text slides.
• Input: The script text from Subagent 1, plus instructions on how to display text (start/end times).
• Output: A structured text timeline (e.g. "0:00–0:04 Hook text," "0:05–0:09 CTA," "0:10–0:14 Hashtags").
SUBAGENT FINAL OUTPUT: [text-overlay-plan]
SUBAGENT 5: "Base Video Creator"
• Skill: #201 - Generate Vertical Text Slide Video From MP3 & Transcription (max 1000 characters)
• Purpose: Takes the chosen .wav file (music) plus the timed overlay text, and creates a vertical text-slide MP4 with background audio.
• Input: One .wav file from Subagent 3, plus the timed text/transcription from Subagent 4.
• Output: A 10–15s vertical MP4 with text slides and background music.
SUBAGENT FINAL OUTPUT: [text-slide-mp4]
SUBAGENT 6: "Final Video Assembler"
• Skill: #199 - Add Images & Videos On Top Of Existing MP4
• Purpose: Layers the b-roll clip on top—or beneath—the text-slide video, combining them into one final short MP4.
• Input: The text-slide MP4 from Subagent 5, the trimmed b-roll MP4 from Subagent 2, and any overlay details needed for positioning.
• Output: The final 10–15s vertical promotional MP4 short video (hook text, CTA, hashtags, background visuals, and music).
SUBAGENT FINAL OUTPUT: [final-promotional-short-mp4]