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 researches the ‘retro 3D anaglyph’ poster style, gathers a concise 300–500 character style summary, locates 3 sample image references, then combines both into a single structured output.”

B) FINAL TASK OUTPUT  
   [concept-and-inspiration-output] → A single structured text (e.g. JSON) containing:  
   • A concise 300–500 character style summary.  
   • Three reference image URLs.

C) SUBAGENT INPUT  
   {variable1} → A text string describing what to research, e.g. "retro 3D anaglyph posters."

E) SUBAGENT TASK SUMMARY

1) #216 – Research Topic (Or Keyword) Deeply  
      • DESCRIPTION: Use {variable1} (e.g. "retro 3D anaglyph posters") to perform deep research.  
      • OUTPUT: A 1000–3000 character result.

2) #190 – Write or rewrite text based on instructions  
      • DESCRIPTION: Rewrite the lengthy research text from #216 down to 300–500 characters (concise style summary).  
      • INPUT: The research output text from #216.  
      • OUTPUT: The final 300–500 character style summary.

3) #187 – Find Relevant Images with Brave Search  
      • DESCRIPTION: Locate 3 references or sample images that illustrate the “retro 3D anaglyph” style.  
      • INPUT: Keyword(s) → “retro 3D anaglyph poster references.”  
      • OUTPUT: 3 image URLs showcasing the style.

4) #223 – Powerful LLM Prompt-to-Text Response  
      • DESCRIPTION: Combine the 300–500 character style summary (from step #190) and the 3 image URLs (from step #187) into one structured output.  
      • INPUT:  
         – Summary from #190  
         – Image URLs from #187  
      • OUTPUT:  
         – [concept-and-inspiration-output] → Single structured text (e.g., JSON or similar) containing the short style summary and references.

F) SILOS  
   • Silo 1: Research & Summarize (Steps #216, #190)  
   • Silo 2: Image References (Step #187)  
   • Silo 3: Final Consolidation (Step #223)

SubAgent #1 - Diagram

Expand Diagram

A) SUBAGENT SUMMARY:  
This subagent (Anaglyph Image Creator) takes the inspiration/concept output from Subagent 1 and uses it to generate one or two “retro 3D anaglyph” style poster images, then returns a consolidated reference to those image URLs.

B) FINAL TASK OUTPUT:  
[anaglyph-designs-output] → A single structured text object (e.g., JSON or list) containing URLs for the 1–2 generated anaglyph images.

C) SUBAGENT INPUT:  
[concept-and-inspiration-output]  
(This would include a style summary and reference images from Subagent 1.)

E) SUBAGENT TASK SUMMARY:

1) The subagent starts by taking in [concept-and-inspiration-output].  
2) It uses Skill #222 (Make Image [Especially With Text]) to generate 1–2 “retro 3D anaglyph” poster images, drawing on the style summary from the input.  
   • INPUT → Text prompt describing the red/cyan anaglyph aesthetic, derived from [concept-and-inspiration-output].  
   • OUTPUT → 1 or 2 URLs of PNG images.  
3) It then uses Skill #223 (Powerful LLM Prompt-to-Text Response) to combine those image URLs into a single structured text reference.  
   • INPUT → Image URLs from #222.  
   • OUTPUT → Single JSON or list referencing the new images.

When complete, the subagent returns [anaglyph-designs-output], which contains the generated image URLs.

F) SILOS:

SILO 1 – Generate Anaglyph Images  
• Step 1 (Skill #222): Input [concept-and-inspiration-output] → Output: 1–2 PNG image URLs

SILO 2 – Consolidate Output  
• Step 2 (Skill #223): Input: PNG URLs from Step 1 → Output: Single structured text referencing all generated images [anaglyph-designs-output]

SubAgent #2 - Diagram

Expand Diagram

A) SUBAGENT SUMMARY  
"Format & Sizing Expert" ensures any input poster images are resized to standard print specs (18×24 inches at 300 DPI, i.e., 5400×7200 px).

B) FINAL TASK OUTPUT  
[print-ready-images-output] → A consolidated reference (e.g., JSON) listing all final 5400×7200 px, 300 DPI image URLs.

C) SUBAGENT INPUT  
[anaglyph-designs-output] → One or more image URLs generated by Subagent 2.

E) SUBAGENT TASK SUMMARY  
1) Receive [anaglyph-designs-output] (containing the 1–2 PNG image URLs).  
2) Skill #191 – Resize Image  
   • Input: Original image URL(s) from [anaglyph-designs-output].  
   • Transformation: Resize each to 5400×7200 px at 300 DPI.  
   • Output: Print-ready PNG URL(s).  
3) Skill #223 – Powerful LLM Prompt-to-Text Response  
   • Input: The list of newly resized URL(s) from #191.  
   • Action: Consolidate them into one structured text format (e.g., JSON array).  
   • Output: [print-ready-images-output] (the final subagent output of print-ready image URLs).

F) SILOS  
• Silo 1: Resize & Format  
   (Skill #191 → returns resized PNG(s))  
• Silo 2: Summarize Output  
   (Skill #223 → returns final structured output).

SubAgent #3 - Diagram

Expand Flow

A) SUBAGENT SUMMARY  
This subagent prepares the final listing metadata (title, short description, up to 5 tags) and combines it with the print‐ready image URLs into a single CSV/JSON snippet.

B) FINAL TASK OUTPUT  
[listing-metadata-output]: A single CSV or JSON file (snippet) containing the product’s title, short description, tags, and any other relevant fields (e.g., image URL(s)).

C) SUBAGENT INPUT  
1) {variable1} – A text prompt or product details describing the “retro 3D anaglyph” poster’s appeal.  
2) [print-ready-images-output] – The final print-ready image URL(s) from the previous subagent.

E) SUBAGENT TASK SUMMARY  
Below is the step-by-step sequence of skills used by this Subagent 4, including inputs/outputs and how they chain together:

1) #190 – Write or rewrite text based on instructions  
   INPUT:  
     • {variable1} (which contains the request to create an SEO-friendly title, short description, and up to 5 tags that highlight the retro 3D anaglyph appeal)  
   OUTPUT:  
     • Text response containing:  
       – Title (brief)  
       – 1–2 sentence description  
       – Up to 5 tags

2) #223 – Powerful LLM Prompt-to-Text Response  
   DESCRIPTION: Format the text results from #190 and include the [print-ready-images-output] URLs in a single CSV or JSON snippet.  
   INPUT:  
     • The title, description, and tags from #190  
     • The image URL(s) from [print-ready-images-output]  
     • Format instructions (to ensure final output is a CSV or JSON snippet)  
   OUTPUT:  
     • [listing-metadata-output] – A final CSV or JSON snippet containing:  
       – Title, description, tags  
       – Print-ready image URL(s)

F) SILOS  
Because this subagent’s objective is straightforward (gather text metadata, merge with image links, and produce final listing info), there is only one silo of tasks:  
• Silo 1: Generate listing text (#190) → Format final CSV/JSON (#223).