What Shall We Build Next?

Describe

Describe your task

Refine

Refine the plan

SubAgents

Review all agents

Deploy

Deploy your agent

A) SUBAGENT SUMMARY  
This subagent’s purpose is to produce a concise, 45-second voice-over script that highlights the course benefits (low cost, only 90 seconds per day, guaranteed 100k views, no need to show your face, 1M+ and even 5M+ real user view counts) while maintaining a fun, semi-serious, flirty/professional tone. It must include a clear call to action directing viewers to facelessflipping.com/a/30.

B) FINAL TASK OUTPUT  
A single text document (≤ 45 seconds read time) containing the finalized voice-over script, ready for narration.

C) SUBAGENT INPUT  
• Key selling points (low cost, 90 seconds/day, 100k minimum views, no face needed, images supplied, real members hitting 1M-5M views)  
• Desired tone (fun, semi-serious, flirty yet professional)  
• Required CTA (facelessflipping.com/a/30)  
• Instruction to keep the script at about 45 seconds total

E) SUBAGENT TASK SUMMARY  
1) Receive the key selling points, tone requirements, and CTA details.  
2) Use Skill #171 - “Write Voice Over Script Based On Instructions” to generate a concise ~45-second script integrating all provided details.  
3) Return the final text script back as the subagent’s final output.

F) SILOS  
Given this subagent has a single primary step, it does not have multiple silos. Its single silo is “Script Generation,” described in the steps above.

SubAgent #1 - Diagram

Expand Diagram

SubAgent #2 - Diagram

Expand Diagram

A) SUBAGENT SUMMARY  
Split the single ~45-second MP3 voice-over into four shorter segments (each corresponding to a different camera angle) and output them as separate audio files.

B) FINAL TASK OUTPUT  
A list of four WAV file URLs (each ~10-12 seconds long), each containing the relevant portion of the original 45-second MP3.

C) SUBAGENT INPUT  
1) The URL of the original 45-second MP3 file.  
2) The start and end timestamps (in seconds) for each of the four desired segments.

D) SUBAGENT TASK SUMMARY  
1) Take the original MP3 URL and the specified start/end timestamps.  
2) Use Skill #219 - “Cut Wav/mp3 Audio into Multiple Pieces/Samples” to cut the MP3 into four segments.  
   • Input to #219:  
       – The original MP3 URL.  
       – The start/end times for each of the four cuts.  
   • Output of #219:  
       – Four separate WAV URLs (one for each segment).  
3) Provide these four WAV files as the final subagent output.

E) SILOS (If Needed)  
Because this subagent has a single, straightforward function—cutting the MP3 from one file into four smaller files—there is effectively just one main “silo” of activity:  
• Audio Splitting Silo  
  – (Skill #219) Cut the single 45-second MP3 into four WAV segments.

No additional subtasks are needed within this subagent. All four audio segments are returned as the final output.

SubAgent #3 - Diagram

Expand Flow

A) SUBAGENT SUMMARY  
Generate four short “talking head” MP4 videos (each from a different camera angle) of a blonde AI newsreader avatar reading out each segment of an MP3 voice-over.

B) FINAL TASK OUTPUT  
An array/list of four MP4 URLs (each 9:16 portrait), one per audio segment, featuring the specified blonde newsreader avatar in different angles, flirty yet professional, dressed in a white blouse with subtle necklace.

C) SUBAGENT INPUT  
• Four MP3 segment URLs (already cut/split from the original 45-second voice-over).  
  (Alternatively, if the subagent is not receiving already-split MP3s, it needs the single 45s MP3 plus time-codes for each segment, but typically that split is done in Subagent 3.)

D) SUBAGENT TASK SUMMARY (Step-by-Step Skill Flow)

1) For each of the four MP3 segment URLs, do the following:  
   (a) Use Skill #198 - “Get Transcription Of MP3 (With Timings)”  
       INPUT: The MP3 segment URL  
       OUTPUT: Transcription text with timestamps  
   (b) Use Skill #168 - “Generate Talking Head Video From MP3 & transcription”  
       INPUT: The same MP3 segment URL + the transcription from step 1(a)  
       NOTES for #168:  
         • Avatar style: blonde hair, shoulder-length, white blouse with some cleavage, subtle necklace, flirty yet professional  
         • 9:16 portrait aspect ratio  
         • Distinguish the camera angle each time (e.g., front, slight left, slight right, over-the-shoulder, etc.)  
       OUTPUT: One MP4 video of the avatar delivering that segment

2) Collect the resulting four MP4 URLs into a single array/list.

3) Output that array/list as [avatar-talking-heads].

E) SILOS  
Because this subagent repeats the same workflow four times (once for each audio segment), you can view them in two mini-silos per segment, repeated 4x:

• Silo 1: Transcription of each segment  
   (Skill #198)  
• Silo 2: Generate talking-head from that transcription & MP3  
   (Skill #168)

Then assemble the four outputs as the final subagent result.

4 Template & Links

Expand Flow

A) SUBAGENT SUMMARY  
“Subagent 5 sequentially combines four separate 9:16 MP4 video clips (each featuring a different camera angle) into one continuous ~40-second MP4 video.”

B) FINAL TASK OUTPUT  
A single MP4 file (“combined-newsreader-angles.mp4”), approximately 40 seconds in duration, containing all four angles stitched together sequentially.

C) SUBAGENT INPUT  
• Four MP4 video URLs (each roughly 10 seconds in length).

D) SUBAGENT TASK SUMMARY  
The subagent takes four distinct MP4 URLs and feeds them into Skill #173 (Concatena Videos), which merges them into a single MP4 video. The output is then saved for use in subsequent steps of the overall workflow.

E) STEP-BY-STEP WORKFLOW  
1) Receive the four finalized avatar MP4 segments (all in 9:16 format).  
2) Call Skill #173 - Concatena Videos (Combine 2-10 MP4s), Must Be Saved.  
   • INPUT: The four avatar MP4 URLs in the desired play order.  
   • OUTPUT: One continuous MP4 video URL (the ~40-second combined newsreader video).  
3) Output the concatenated MP4 URL as [combined-newsreader-angles].

F) SILOS  
No additional silos are required for this subagent. The entire subagent is one straightforward silo:  
• Combine all four MP4s (Skill #173) → Produce single combined MP4.

5 Template & Links

Expand Flow

A) SUBAGENT SUMMARY  
This subagent creates a short five-second animated clip (in 9:16 format) that showcases the main bullet points (e.g., “Only 90 seconds daily,” “Guaranty of 100k views or money back,” etc.) and finishes with the “facelessflipping.com/a/30” call to action.

B) FINAL TASK OUTPUT  
A single 5-second MP4 animated video (vertical 9:16) featuring the key bullet points and final CTA, labeled as [5s-animated-clip].

C) SUBAGENT INPUT  
• Relevant bullet points or short phrases to highlight (e.g., “Only 90 seconds daily,” “Guaranteed 100k+ views,” “No face needed,” “1M+ or 5M+ views possible,” etc.).  
• CTA URL: “facelessflipping.com/a/30.”  
• Optional brand elements or color/style preferences, if desired.

E) SUBAGENT TASK SUMMARY  
Below is the step-by-step workflow for this subagent:

1) (Optional) Format/Create Prompt Text (if needed)  
   – Skill #190 (Write or rewrite text based on instructions)  
   – Purpose: Take the subagent’s bullet points/CTA and generate a concise text prompt describing the desired animation style, aspect ratio (9:16), and final CTA.  
   – Input: The bullet points/list of features, plus any style/branding notes.  
   – Output: A short text prompt suitable for the animation skill.

2) Generate 5-Second Animated Clip  
   – Skill #169 (Create A 5 Second Animated Video From Text Prompt)  
   – Purpose: Take the final text prompt and produce the 5-second MP4 animation in 9:16.  
   – Input: The prompt from step 1 (or the directly provided bullet points + CTA, if no rewriting is needed).  
   – Output: A 5-second MP4 animated clip featuring the bullet points and displaying the CTA at the end.  
   – Final Output for the Subagent: [5s-animated-clip].

F) SILOS  
This subagent is small enough that it effectively functions in a single silo:  
• Silo 1: Prompt preparation (optional) → 5-second animation generation

No further silos are required. The final MP4 (9:16, 5 seconds) completes the subagent’s objective.