AI Prompt Architect & Workflow Guide

Transform basic ideas into engineered prompts and identify high-value AI workflows.

🛠️ Prompt Architect Tool

🚀 Optimized Prompt


            


            

                
🤖 Recommended Model & Tool

                

            

        

    


    
    

        
📚 Glossary of Concepts

        

            

                Triple-Check Prompting
                A rigorous framework that forces the AI to analyze variables, criticize its own draft, and adopt multiple expert perspectives before finalizing an answer.
            

            

                Few-Shot Prompting
                Providing the AI with a few examples (shots) of the input and desired output to set a specific pattern or style.
            

            

                Chain-of-Thought (CoT)
                Instructing the AI to "think step-by-step" or show its reasoning process, which significantly reduces errors in logic and math.
            

            

                Tree-of-Thought (ToT)
                Asking the AI to simulate multiple experts exploring different "branches" of solutions and voting on the best one.
            

            

                Prompt Chaining
                Breaking a complex task into a sequence of smaller prompts, where the output of one becomes the input for the next.
            

            

                RAG (Retrieval-Augmented Generation)
                Grounding the AI's response in specific, uploaded documents (Knowledge) rather than its general training data.
            

            

                Bot Stacking
                A BoodleBox feature where you use different AI models (e.g., Perplexity for research, Claude for writing) in the same chat context.
            

            

                Nano Banana
                Google's Gemini Flash Image generator. Fast, watermark-free, and ideal for creating educational visuals.
            

            

                Star Docs
                Documents you "star" in BoodleBox to automatically attach them to every new chat (e.g., Style Guides, Policies).
            

        

    


    
    

        
Identifying AI Workflows: Two Strategic Models

        
Not sure what to prompt for? Use these two models to identify high-impact opportunities in your organization.


        

            
            

                
Problem-Based Learning (PBL)

                
Focus: Identifying known pain points and challenges that can be solved more intelligently with AI. This model starts with the "Why" and focuses on measurable value and ROI.

                
                
🏫 K-16 Education Examples

                
    
                    
  • 
                        Problem: IEP paperwork takes 15+ hours/week.
                        
    Value: Reduce drafting time by 50%, saving $200k in turnover costs.
    
                    
    • 
                    
  • 
                        Problem: Feedback on writing takes 2 weeks to return.
                        
    Value: Instant AI-assisted feedback loops improve proficiency scores by 15%.
    
                    
    • 
                


                
🏢 Organization Examples

                
    
                    
  • 
                        Problem: Low donor retention due to generic "blast" emails.
                        
    Value: Personalized AI impact reports increase retention by 20%.
    
                    
    • 
                    
  • 
                        Problem: Directors spend 40% of time on manual report formatting.
                        
    Value: Automating "raw data to presentation" recovers 15 hours/week per director.
    
                    
    • 
                

            


            
            

                
Use Case Model

                
Focus: Identifying "quick-win" pilot projects with high success probabilities. This model unbundles jobs into specific tasks based on four identifying characteristics.

                
                
The Four Pillars

                

                    Data-Driven
                    Repetitive
                    Predictive
                    Generative
                


                
🏫 K-16 Education Examples

                
    
                    
  • Generative: Creating watermark-free visual aids or diagrams via Nano Banana.
    • 
                    
  • Repetitive: Using Custom Bots to answer Syllabus/Handbook FAQs 24/7.
    • 
                    
  • Data-Driven: Analyzing CSV test scores via Code Interpreter to identify learning gaps.
    • 
                


                
🏢 Organization Examples

                
    
                    
  • Repetitive: Automating meeting minutes using Star Docs and a style guide.
    • 
                    
  • Generative: Repurposing one grant report into a month of social content via Bot Stacking.
    • 
                    
  • Data-Driven: Creating vendor comparison matrices from multiple PDF proposals.
    •