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 is responsible for creating a structured high-level course outline (with modules, objectives, and prerequisites) on any given topic—in this case, “Smart Home Automation.”

B) FINAL TASK OUTPUT  
A single text-based course outline (modules, objectives, prerequisites) in plain text format (≤ 10,000 characters). This text output is referred to as [course-outline].

C) SUBAGENT INPUT  
• A short text prompt specifying the topic to outline (e.g. “Smart Home Automation”)

E) SUBAGENT TASK SUMMARY  
Below is the step-by-step flow of how this subagent will operate.

1) Skill #218 – Brainstorm Related Keywords  
   INPUT:  
      – The initial topic prompt (e.g. “Smart Home Automation”).  
   OUTPUT:  
      – A list of up to 10 related subtopics/keywords.

2) Skill #216 – Research Topic (Or Keyword) Deeply  
   INPUT:  
      – The original topic plus any relevant subtopics gathered from Skill #218 (e.g., “Smart Home Automation + top related keywords”)  
   OUTPUT:  
      – A ~1000 to 3000-character research summary covering the deeper context and core concepts of the topic.

3) Skill #223 – Powerful LLM Prompt-to-Text Response  
   INPUT:  
      – The full research summary from Skill #216, plus a prompt instructing the model to produce a course outline with modules, prerequisites, and objectives for the topic.  
   OUTPUT:  
      – A comprehensive text-based course outline (the final subagent output).

F) SILOS (Optional Organization)  
1) Silo A: Preliminary Keyword Brainstorm  
   – Use Skill #218 with the input topic to gather the most relevant keywords and subtopics.

2) Silo B: Deep Research  
   – Use Skill #216, feeding in the original topic and the subtopics from Silo A to generate a cohesive research summary.

3) Silo C: Outline Generation  
   – Use Skill #223 on the research summary to produce the final text-based course outline.

–––––––––––––––––––––––––––––––––––––––––––––––––––  
SUBAGENT FINAL OUTPUT: [course-outline]

SubAgent #1 - Diagram

Expand Diagram

A) SUBAGENT SUMMARY  
This subagent (Lesson Content Writer) is responsible for producing detailed lesson text for each module in the “Smart Home Automation” course, including examples, use cases, best practices, and relevant imagery (both found online and custom-generated).

B) FINAL TASK OUTPUT  
The subagent’s final output is [lesson-text-and-images], which will consist of:  
• A coherent block of written lesson text for each course module (structured, with examples).  
• Up to five relevant image URLs gathered via Brave Search for each lesson.  
• Any custom generated diagrams or labeled images (PNG URLs).

C) SUBAGENT INPUT  
1. The course outline (with modules, subtopics, etc.).  
2. Any specific formatting or thematic instructions for the lessons.  
3. Any instructions on image styles, constraints, or guidelines for custom diagrams.

E) SUBAGENT TASK SUMMARY  
Below is the detailed workflow of how the subagent processes each module to produce the lesson text and images:

1) Receive Outline & Identify Module(s):  
   - The subagent takes in the course outline, containing the module list and relevant instructions.

2) For Each Module (repeat steps 2.1 – 2.4 as necessary):

2.1) RESEARCH MODULE USING #216 - Research Topic (Or Keyword) Deeply  
       INPUT:  
         • The module title or topic name (e.g., “Smart Lighting,” “Home Security Integration”).  
       OUTPUT:  
         • A 1000–3000 character summary of the researched content, focusing on relevant technology, examples, best practices, and subtopics.

2.2) WRITE MODULE LESSON TEXT USING #223 - Powerful LLM Prompt-to-Text Response  
       INPUT:  
         • The research summary from #216.  
         • Instructions to convert that summary into a structured lesson (including definitions, how-to, use cases, benefits, pitfalls, etc.).  
       OUTPUT:  
         • A cohesive, well-structured lesson text for the module, ready to be integrated into the course.

2.3) FIND RELEVANT IMAGES USING #187 - Find Relevant Images with Brave Search  
       INPUT:  
         • Keywords or topics from the module (e.g., “smart thermostat,” “Z-Wave diagram,” “voice assistant device,” etc.).  
       OUTPUT:  
         • A list of up to 5 URLs of relevant images.

2.4) CREATE ANY CUSTOM DIAGRAMS USING #222 - Make Image (Especially With Text)  
       INPUT:  
         • A prompt describing the needed diagram or labeled illustration (e.g., “Diagram showing how a smart hub controls multiple IoT devices in a home network”).  
       OUTPUT:  
         • A PNG image URL containing the custom diagram.

3) Compile Results for Each Module:  
   - Collect the completed lesson text from step 2.2, relevant image URLs from step 2.3, and any custom-generated diagrams from step 2.4.  
   - Move on to the next module until all modules are processed.

4) Produce Final Subagent Output:  
   - Combine all the module lesson texts and associated images (including custom diagrams) into one structured result: [lesson-text-and-images].

F) SILOS  
SILO #1: “Module-by-Module Content Generation”  
• Steps:  
  1. Research module (#216).  
  2. Draft lesson text (#223).  
  3. Find relevant images (#187).  
  4. Optionally create custom diagrams (#222).  
• Repeats for each module in the outline.

Upon completion, subagent 2 assembles the final [lesson-text-and-images] output, ready for integration into the full course.

SubAgent #2 - Diagram

Expand Diagram

A) SUBAGENT SUMMARY  
This subagent will generate quizzes for each lesson module, based on the final lesson text (from Subagent 2), returning a consolidated set of review questions with optional answers/hints.

B) FINAL TASK OUTPUT  
A single text-based output (e.g., “[quiz-questions]”) containing a curated set of quiz items (multiple-choice, short answer, T/F, etc.) for each module.

C) SUBAGENT INPUT  
1) The complete lesson text from Subagent 2, ideally organized per module.  
2) Instructions specifying the type and number of questions (e.g., “5 multiple-choice, 2 true-false, 3 short-answer”).

E) SUBAGENT TASK SUMMARY  
1) Receive “final lesson text” (covering all modules) and “instructions” on quiz format.  
2) For each module in the lesson text:  
   • Use Skill #223 (Powerful LLM Prompt-to-Text Response) to generate quiz questions, passing in (a) the module’s lesson text and (b) the question-type/quantity instructions.  
   • Collect the generated quiz questions (and optional hints/answers).  
3) Combine (append) all module quizzes into one final text output (optionally using Skill #190 - Write or rewrite text based on instructions - if we need to unify style or formatting).  
4) Provide final consolidated quiz text or JSON structure as “[quiz-questions].”

F) SILOS

• Silo 1: “Module Quiz Generation”  
  - For each module’s lesson text →  
    → Skill #223: Generate quiz questions based on instructions (types/quantity).  
  - Output: Text containing quiz items for that module.

• Silo 2: “Aggregation & Final Output”  
  - Combine/format all quizzes (optionally via Skill #190 if specific formatting is needed).  
  - Return final quiz compilation as “[quiz-questions].”

SubAgent #3 - Diagram

Expand Flow

A) SUBAGENT SUMMARY:  
This subagent takes all the text-based course components (outline, lessons, quizzes) plus references to images and weaves them together into one properly formatted, combined document, ready for external PDF generation.

B) FINAL TASK OUTPUT:  
A single, final “combined” text (or HTML) file containing all course content (outline, lessons, quiz questions) along with references/placement instructions for images, suitable for converting to PDF.

C) SUBAGENT INPUT:  
• The course outline text (from Subagent 1)  
• The lesson texts (from Subagent 2)  
• Quiz questions (from Subagent 3)  
• Image references/URLs or any custom graphics references (from Subagent 2)  
• Formatting guidance (e.g., headings, subheadings, placement, style instructions)

D) SUBAGENT TASK SUMMARY (Step-by-step skill usage):  
1) Gather All Inputs  
   - Collect the outline, lesson texts, quiz questions, and image references/URLs in a single prompt or structure.  
2) Use Skill #190 - “Write or rewrite text based on instructions”  
   • INPUT to Skill #190:  
     - A prompt containing:  
       1) All of the text from outline, lessons, and quizzes.  
       2) References to images/URLs.  
       3) Detailed instructions on how to format the final document (e.g., headings, subheadings, where to place images, page breaks if needed).  
   • OUTPUT from Skill #190:  
     - A single, combined text (or HTML) document that merges all course components in a logical, well-structured format (e.g., with placeholders where images should go, headings, subheadings, etc.).  
3) Subagent Final Output  
   • The consolidated text or HTML document ([final-combined-document]) is returned.  
   • This can then be passed to an external PDF generator (outside this skill set).

E) SILOS:  
• Silo 1: Collect & Organize Inputs  
   - Collate outline, lessons, images, and quiz content into an organized data structure (text blocks + image references).  
• Silo 2: Single Rewrite & Formatting Step  
   - Invoke Skill #190 with all compiled content and instructions to produce the final, combined output.

That completes the subagent flow for compiling and formatting the PDF course content.

4 Template & Links

Expand Flow

A) SUBAGENT SUMMARY:  
Subagent #5 is responsible for creating a set of video lessons (one per module) for the Smart Home Automation course by using the finalized text content from previous subagents, turning that text into voice-over MP3 files, and then generating text-slide videos. Finally, it compiles these individual lesson videos into one complete course video.

B) FINAL TASK OUTPUT:  
A single MP4 video file (16:9 aspect ratio) containing all lesson modules of the Smart Home Automation course, each with synchronized text slides and narrated voice-over.

C) SUBAGENT INPUT:  
1. The final text for each lesson module (produced by previous subagents).  
2. Instructions regarding the style of the video slides (e.g., branding guidelines, text size, color scheme).  
3. Optional: any images or graphics to overlay on the video.

D) SUBAGENT TASK SUMMARY (Step-by-Step):  
1. Receive final lesson text for each module.  
2. For each module:  
   • Generate a voice-over script if needed (#171) – optional if the lesson text is already in a suitable script format.  
   • Convert the final script/lesson text to an MP3 voice-over (#170).  
   • Generate a text-slide video using the MP3 voice-over and transcript (#200).  
   • (Optional) Overlay additional images on the video (#199).  
3. Concatenate all module videos into a single final MP4 file (#173).  
4. Output the final MP4 URL.

E) SILOS:

• SILO 1: VOICE OVER PREPARATION  
  (a) Skill #171 - Write Voice Over Script Based On Instructions (if needed)  
      INPUT: Lesson text + any specific instructions for a voice-over script format.  
      OUTPUT: A short script (100-2000 words).  
  (b) Skill #170 - Turn Script Into Voice Over MP3  
      INPUT: The voice-over script text.  
      OUTPUT: An MP3 file URL (voice-over).

• SILO 2: VIDEO CREATION (PER MODULE)  
  (a) Skill #200 - Generate Text Slide Video From MP3 & Transcription (max 2000 characters)  
      INPUT: MP3 URL (from Silo 1) + transcript text (final lesson text or voice-over script).  
      OUTPUT: Module-level MP4 video (text slides synced to voice-over).  
  (b) (Optional) Skill #199 - Add Images & Videos On Top Of Existing MP4  
      INPUT: The newly created MP4 from skill #200 + images or short video clips + (optional) transcript for context.  
      OUTPUT: Enhanced MP4 video with inserted/overlaid images or videos.

• SILO 3: VIDEO COMPILATION  
  Skill #173 - Concatena Videos (Combine 2-10 MP4s)  
    INPUT: The MP4 lesson videos from Silo 2 (one per module).  
    OUTPUT: Single concatenated MP4 containing all lesson modules.

This completes Subagent #5’s workflow for generating a combined video course on Smart Home Automation.