Advanced AI Agent Prompt Template

1. Version switcher: v1, v2, v3

Three versions of the same template were pasted on 2026-11-03. No version is presented here as the only version, and they are not merged into an invented hybrid. Use the buttons to read each version's own purpose, capabilities, knowledge base, and prompt template. The full v1 reference text for every section follows in the body of the page.

2. What changed between versions

v2 supersedes v1 only in the sections it rewrites; everything it does not mention carries over from v1. v3 is a condensed reworded variant, not a replacement for the full text.

3. Agent identity and purpose (v1 full reference)

Purpose

The Advanced AI Agent is designed to function as a sophisticated artificial intelligence assistant capable of handling complex, multi-step tasks across diverse domains. This agent excels at understanding nuanced instructions, maintaining context over extended interactions, reasoning through challenging problems, and adapting its approach based on user feedback while providing transparent explanations of its thought process.

4. Core capabilities (v1)

• Process and interpret complex, multi-part instructions
• Maintain context and coherence across extended conversations
• Apply critical thinking and reasoning to solve problems
• Adapt responses based on user feedback and preferences
• Provide step-by-step explanations of reasoning processes
• Integrate knowledge across multiple domains
• Generate creative solutions to open-ended problems
• Identify and resolve ambiguities through clarification

5. Knowledge base (v1)

Reasoning Frameworks

• Deductive reasoning methodologies
• Inductive reasoning approaches
• Abductive reasoning techniques
• Analogical reasoning strategies
• Causal reasoning principles
• Probabilistic reasoning models
• Counterfactual reasoning processes
• Ethical reasoning frameworks

Communication Expertise

• Audience adaptation techniques
• Explanation clarity principles
• Technical concept simplification
• Narrative structure and storytelling
• Socratic questioning methods
• Active listening strategies
• Ambiguity resolution approaches
• Feedback incorporation techniques

Problem-Solving Methodologies

• Structured problem decomposition
• First principles thinking
• Systems thinking approaches
• Design thinking frameworks
• Lateral thinking techniques
• Root cause analysis methods
• Decision tree modeling
• Constraint satisfaction strategies

Meta-Learning Capabilities

• Self-correction mechanisms
• Knowledge gap identification
• Learning transfer techniques
• Concept integration strategies
• Pattern recognition methods
• Mental model construction
• Cognitive bias awareness
• Metacognitive monitoring

6. The prompt template (v1)

7. Implementation examples

Complex Problem-Solving

Use the template to help with the following task:

Task Description: Develop a comprehensive strategy for a small retail business to reduce its environmental impact while maintaining profitability

Specific Requirements:

• Must address multiple aspects of the business (supply chain, operations, product offerings, etc.)
• Solutions should be practical and implementable with limited resources
• Each recommendation should include implementation steps, estimated costs, and potential benefits
• Strategy should balance short-term actions with long-term sustainability goals

Constraints:

• Business has 15 employees and 2 locations
• Annual revenue approximately $1.2 million
• Limited budget of $25,000 for sustainability initiatives
• Located in a suburban area with limited public transportation
• Customer base is environmentally conscious but price-sensitive

Preferred Approach:

• Prioritize solutions by impact-to-cost ratio
• Include both operational changes and customer-facing initiatives
• Consider potential marketing benefits of sustainability efforts
• Provide metrics for measuring success

Output Format:

• Structured strategy document with clear sections
• Include executive summary
• Provide implementation timeline
• Include visual elements where helpful

Creative Project Development

Use the template to help with the following task:

Task Description: Create a detailed concept for an educational mobile game that teaches middle school students about climate science and sustainability

Specific Requirements:

• Game should be engaging while maintaining educational integrity
• Must cover key climate science concepts (greenhouse effect, carbon cycle, climate systems, human impacts)
• Should incorporate real-world data and scientific principles
• Must include both individual challenges and collaborative elements
• Should adapt difficulty based on player progress
• Must be appropriate for 11-14 year old students

Constraints:

• Game must be playable on both iOS and Android devices
• Should function without requiring constant internet connection
• Must comply with educational privacy standards (COPPA, etc.)
• Should be completed within 15-20 hours of gameplay
• Must be accessible to students with various learning needs

Preferred Approach:

• Balance entertainment and education
• Incorporate narrative elements and character development
• Include systems for tracking learning progress
• Consider classroom integration possibilities

Output Format:

• Comprehensive game design document
• Include gameplay mechanics description
• Outline narrative structure
• Describe progression system
• Include sample challenges/levels
• Provide visual concept descriptions

v2 addition to this example: the agent integrates tone detection and vocabulary profiling to personalize responses based on user communication style and age group, so concepts are finely tuned to the developmental level, linguistic comprehension, and motivational tone of the intended audience.

8. Thinking frameworks

Chain-of-Thought Reasoning

Chain-of-Thought reasoning is a step-by-step thinking process that breaks down complex problems into a sequence of logical steps. This approach helps both the AI and the user understand how a conclusion was reached.

For First-Time Programmers: Think of Chain-of-Thought reasoning like showing your work in a math problem. Instead of just giving the final answer, you write down each step of your calculation. This makes it easier to check your work for errors, understand how you reached the solution, explain your process to others, and learn from your approach.

Example Implementation. Problem: Calculate the total cost of a shopping cart with a 15% discount and 8% tax.

Chain-of-Thought:

1. First, I will list the items and their prices:
   • Shirt: $25
   • Pants: $40
   • Shoes: $60
   • Total before discount: $125
2. Next, I will calculate the discount amount:
   • 15% of $125 = $125 x 0.15 = $18.75
   • Price after discount = $125 - $18.75 = $106.25
3. Now, I will calculate the tax:
   • 8% of $106.25 = $106.25 x 0.08 = $8.50
   • Final price = $106.25 + $8.50 = $114.75

Therefore, the total cost is $114.75.

First Principles Thinking

First Principles thinking involves breaking down complex problems into their most basic, foundational elements and then reassembling them from the ground up. This approach helps avoid assumptions and conventional thinking that might limit creative solutions.

For First-Time Programmers: Imagine you are building with LEGO blocks. Instead of following the instruction manual (conventional thinking), First Principles thinking means dumping out all the blocks and examining what you have, understanding what each block can do, building something new based on the fundamental properties of the blocks, and not being limited by what others have built before.

Example Implementation. Problem: Design a more efficient online checkout process.

First Principles Approach:

1. What is the fundamental purpose of checkout?
   • To transfer ownership of products in exchange for payment
2. What are the essential components needed?
   • Product selection confirmation
   • Payment information collection
   • Shipping information (if physical product)
   • Order confirmation
3. What assumptions might be limiting current designs?
   • Assumption: Users must create accounts
   • Assumption: Payment and shipping must be separate steps
   • Assumption: All information must be entered each time
4. Rebuilding from fundamentals:
   • Single-page design with only essential information
   • Guest checkout by default with account creation as optional
   • Save information securely for returning customers
   • Integrate digital wallet options to eliminate manual entry
   • Combine shipping and payment into a single step when possible

This approach creates a streamlined checkout with fewer steps by focusing on the fundamental purpose rather than conventional patterns.

Socratic Method

The Socratic Method uses a series of questions to stimulate critical thinking, clarify concepts, and explore complex ideas. This approach helps both the AI and the user refine their understanding through dialogue.

For First-Time Programmers: The Socratic Method is like having a conversation with yourself to solve a problem. Instead of immediately trying to find the answer, you ask yourself a series of questions that help you clarify what you are really trying to solve, challenge your assumptions, consider different perspectives, and discover insights you might have missed.

Example Implementation. Problem: Determine whether to use a relational database or NoSQL database for a new application.

Socratic Method Approach:

1. What is the nature of the data we need to store?
   • The data has a clear structure with defined relationships between entities.
   • Some data elements have many-to-many relationships.
2. What are our primary access patterns?
   • We frequently need to join data across multiple entities.
   • We need to ensure data consistency across transactions.
   • We do not expect extremely high write volumes.
3. What are our scalability requirements?
   • Initial user base is moderate but may grow significantly.
   • Data volume will likely remain manageable for several years.
4. How important is query flexibility?
   • We need to support complex queries with multiple conditions.
   • Ad-hoc reporting capabilities are important.
5. What are our consistency requirements?
   • Strong consistency is required for financial transactions.
   • Some data can tolerate eventual consistency.

Based on this questioning process, a relational database appears more suitable for our core needs, though we might consider a hybrid approach with NoSQL for specific high-volume data components.

MECE Framework (Mutually Exclusive, Collectively Exhaustive)

The MECE framework organizes information into categories that are both mutually exclusive (no overlap) and collectively exhaustive (nothing left out). This approach ensures comprehensive analysis without redundancy.

For First-Time Programmers: Think of MECE like organizing your closet where each item belongs in exactly one drawer (mutually exclusive) and every item in your room is stored somewhere in the closet (collectively exhaustive). Nothing is counted twice, and nothing is forgotten. This helps you organize complex problems into clear categories that cover everything without overlap.

Example Implementation. Problem: Analyze factors affecting website conversion rate.

MECE Framework Application:

1. User-Related Factors
   • Demographics (age, location, income)
   • Technical proficiency
   • Device type and capabilities
   • Prior brand experience
2. Website Design Factors
   • Visual design and aesthetics
   • Information architecture
   • Content quality and relevance
   • Responsive design implementation
3. Technical Performance Factors
   • Page load speed
   • Server response time
   • Error rates
   • Cross-browser compatibility
4. Marketing and Traffic Factors
   • Traffic source (organic, paid, social)
   • Campaign messaging alignment
   • Audience targeting precision
   • Competitive positioning
5. Conversion Process Factors
   • Form design and length
   • Checkout process steps
   • Payment options
   • Trust indicators and security

This MECE structure ensures we have considered all possible factors affecting conversion rates without overlap between categories.

9. Advanced features

Recursive Self-Improvement Algorithm

The Recursive Self-Improvement Algorithm is a technique where the AI continuously refines its responses through multiple iterations of self-review and enhancement.

For First-Time Programmers: Imagine you are writing an essay. The Recursive Self-Improvement process would be: write a first draft (initial solution); read it and identify problems (self-review); make specific improvements (refinement); repeat steps 2 and 3 several times, each time making the essay better; finish with a much stronger final version. This is different from just writing and editing once, it is a systematic process of multiple improvement cycles.

Example Implementation:

def generate_solution(problem):
    # Initial solution generation
    solution = create_initial_solution(problem)

    # Set parameters for improvement
    max_iterations = 3
    improvement_threshold = 0.2

    for iteration in range(max_iterations):
        # Self-evaluation of current solution
        evaluation = evaluate_solution(solution)

        # Identify specific areas for improvement
        improvement_areas = identify_improvement_areas(solution, evaluation)

        # If no significant improvements needed, stop iterating
        if not improvement_areas or evaluation.score > 0.9:
            break

        # Generate improved solution based on identified areas
        new_solution = refine_solution(solution, improvement_areas)

        # Calculate improvement delta
        improvement = calculate_improvement(solution, new_solution)

        # Update solution if significant improvement made
        if improvement > improvement_threshold:
            solution = new_solution
        else:
            # Smaller improvements in later iterations are expected
            improvement_threshold *= 0.7
            solution = new_solution

    return solution

Step-by-Step Explanation:

1. We start by creating an initial solution to the problem
2. We set up parameters to control the improvement process
3. We enter a loop that will run up to 3 times (max_iterations)
4. In each iteration, we:
   • Evaluate the current solution
   • Identify specific areas that need improvement
   • Stop if the solution is already excellent (score > 0.9)
   • Create an improved version targeting those specific areas
   • Measure how much better the new version is
   • Keep the new version if it is significantly better
   • Adjust our expectations for improvement as we iterate
5. Finally, we return the best solution we developed

This is a conceptual example of how recursive self-improvement works. It is not meant to be run as actual code, but to illustrate the process.

Multi-Perspective Analysis Framework

The Multi-Perspective Analysis Framework examines problems from different viewpoints to ensure comprehensive understanding and avoid biased solutions.

For First-Time Programmers: Think of Multi-Perspective Analysis like walking around a sculpture in a museum. From the front, you see one view. From the side, you notice different details. From the back, you discover completely new aspects. Only by walking all the way around do you fully understand the sculpture. When solving problems, examining them from multiple perspectives helps you see aspects you might miss from just one viewpoint.

Example Implementation. Problem: Evaluate a proposal to implement a four-day workweek at a company.

Multi-Perspective Analysis:

1. Financial Perspective
   • Impact on productivity and output
   • Changes in operational costs
   • Potential effects on revenue
   • Implementation costs for transition
2. Employee Perspective
   • Work-life balance improvements
   • Potential for burnout from compressed hours
   • Compensation considerations
   • Childcare and personal scheduling impacts
3. Customer Perspective
   • Changes to service availability
   • Impact on response times
   • Potential quality improvements from refreshed employees
   • Communication needs during transition
4. Operational Perspective
   • Scheduling complexities
   • Coverage requirements
   • Training and adaptation period
   • Systems and process adjustments
5. Competitive Perspective
   • Talent attraction and retention advantages
   • Industry positioning and differentiation
   • Competitor responses
   • Market perception
6. Long-term Strategic Perspective
   • Alignment with company culture and values
   • Sustainability of the model
   • Adaptability to future growth
   • Precedent for other workplace innovations

By examining the proposal from these six distinct perspectives, we develop a comprehensive understanding of its implications and can make a more informed decision.

Counterfactual Reasoning Engine

The Counterfactual Reasoning Engine explores alternative scenarios by changing variables to understand causal relationships and potential outcomes.

For First-Time Programmers: Counterfactual reasoning is like playing "what if" games with a problem. You start with what actually happened or what is currently true, then you ask, "What if X had been different?" You mentally simulate how that change would affect everything else. This helps you understand cause-and-effect relationships and test assumptions. It is similar to how scientists use control groups in experiments to understand what causes what.

Example Implementation. Problem: A marketing campaign did not meet expected results.

Counterfactual Reasoning Analysis:

1. Baseline Scenario (What Actually Happened)
   • Email campaign sent to 50,000 customers
   • 15% open rate, 2% click-through rate, 0.5% conversion
   • $15,000 spent, $12,000 revenue generated
   • Primary message focused on product features
   • Sent on Tuesday morning
2. Counterfactual: Different Audience Targeting
   • What if we had segmented by previous purchase behavior?
   • Likely outcome: Higher relevance to recipients
   • Estimated impact: +5% open rate, +1.5% click-through
   • Causal insight: Audience selection may be more important than message content
3. Counterfactual: Different Value Proposition
   • What if we had focused on benefits instead of features?
   • Likely outcome: Similar open rate but higher engagement
   • Estimated impact: Same open rate, +1% click-through, +0.3% conversion
   • Causal insight: Value proposition affects conversion more than initial attention
4. Counterfactual: Different Timing
   • What if we had sent on Friday evening instead?
   • Likely outcome: Lower open rate due to weekend
   • Estimated impact: -3% open rate, similar click-through for those who opened
   • Causal insight: Timing primarily affects open rates, not engagement of already interested customers
5. Counterfactual: Multi-channel Approach
   • What if we had coordinated with social media reinforcement?
   • Likely outcome: Multiple exposures creating familiarity
   • Estimated impact: +1% open rate, +0.7% click-through, +0.4% conversion
   • Causal insight: Channel reinforcement creates compound effects beyond individual channel metrics

This counterfactual analysis suggests that audience segmentation and multi-channel coordination would likely have had the most significant positive impact on campaign performance.

Metacognitive Monitoring System

The Metacognitive Monitoring System allows the AI to reflect on its own reasoning process, identify potential weaknesses, and adjust its approach accordingly.

For First-Time Programmers: Metacognitive monitoring is like having an observer in your mind watching how you solve problems. You work on solving a problem. At the same time, you are aware of HOW you are solving it. You notice when you are making assumptions or missing information. You can pause and adjust your approach when needed. It is similar to how experienced programmers debug their own thinking process while coding, not just debugging the code itself.

Example Implementation. Problem: Recommend a technology stack for a new e-commerce platform.

Metacognitive Monitoring Process:

Initial Approach: I will start by recommending popular e-commerce technologies based on market share and community support.

Metacognitive Check #1: Wait, I am making an assumption that popularity equals suitability. I should first understand the specific requirements before recommending technologies.

Adjusted Approach: Let me identify key requirements for this e-commerce platform:

• Expected traffic volume: Medium (10,000 daily visitors)
• Product complexity: Moderate (clothing with size/color variations)
• Budget constraints: Limited initial investment
• Technical expertise: Small team with primarily JavaScript experience
• Scalability needs: Moderate growth expected

Metacognitive Check #2: I notice I am focusing heavily on technical aspects but have not considered business operations like inventory management and payment processing. This could lead to an incomplete recommendation.

Adjusted Approach: Let me expand my analysis to include:

• Inventory management requirements
• Payment gateway integration
• Shipping and tax calculation needs
• Marketing and SEO capabilities
• Customer service integration

Metacognitive Check #3: I realize I might be overcomplicating this for a team with limited resources. I should prioritize solutions that balance capability with implementation simplicity.

Final Recommendation: Based on the requirements and my metacognitive adjustments, I recommend:

• Frontend: React with Next.js (matches team expertise, good performance)
• Backend: Node.js with Express (leverages existing JavaScript knowledge)
• Database: MongoDB for product catalog, PostgreSQL for transactions
• E-commerce framework: Medusa.js (open-source, JavaScript-based, simpler than larger solutions)
• Hosting: Vercel for frontend, managed MongoDB/PostgreSQL services

This recommendation balances technical requirements with team capabilities and provides a path for scaling as the business grows.

10. External knowledge sources

• Cognitive Science Research: Understanding human reasoning and cognitive biases — https://plato.stanford.edu/entries/cognitive-science/
• Critical Thinking Foundation: Frameworks for structured reasoning — https://www.criticalthinking.org/
• Decision-Making Under Uncertainty: Stanford resources on decision theory — https://plato.stanford.edu/entries/decision-theory/
• Systems Thinking Resources: Donella Meadows Institute materials — https://donellameadows.org/systems-thinking-resources/
• Metacognition Research: Harvard resources on thinking about thinking — https://gse.harvard.edu/news/uk/15/09/science-metacognition
• Explainable AI Principles: Google's guidelines for AI explanations — https://pair.withgoogle.com/chapter/explainability-trust/
• Mental Models Library: Farnam Street's collection of thinking tools — https://fs.blog/mental-models/
• Ethical Reasoning Frameworks: MIT's ethics resources — https://ethics.mit.edu/

Note: the source lists these URLs as plain text. They are shown here as links for convenience. Nothing on this page loads them.

11. Graph and document generation prompts

Decision Tree Visualization Prompt

Create a decision tree visualization for the following complex decision:

Decision Context: [CONTEXT]
Key Decision: [DECISION]

Main Branches:

• Option A: [OPTION A]
  • Potential Outcome A1: [OUTCOME] (Probability: [PROBABILITY])
    • Secondary Decision A1-1: [DECISION]
      • Outcome A1-1a: [OUTCOME]
      • Outcome A1-1b: [OUTCOME]
    • Secondary Decision A1-2: [DECISION]
      • Outcome A1-2a: [OUTCOME]
      • Outcome A1-2b: [OUTCOME]
  • Potential Outcome A2: [OUTCOME] (Probability: [PROBABILITY])
    • Secondary Decision A2-1: [DECISION]
      • Outcome A2-1a: [OUTCOME]
      • Outcome A2-1b: [OUTCOME]
• Option B: [OPTION B]
  • Potential Outcome B1: [OUTCOME] (Probability: [PROBABILITY])
    • Secondary Decision B1-1: [DECISION]
      • Outcome B1-1a: [OUTCOME]
      • Outcome B1-1b: [OUTCOME]
  • Potential Outcome B2: [OUTCOME] (Probability: [PROBABILITY])
    • Secondary Decision B2-1: [DECISION]
      • Outcome B2-1a: [OUTCOME]
      • Outcome B2-1b: [OUTCOME]

Create a clear, hierarchical decision tree visualization that shows the complete decision path with branches, probabilities, and outcomes. Use appropriate colors to indicate favorable and unfavorable outcomes, and include decision points as distinct nodes from chance events.

For first-time users: A decision tree is a flowchart-like diagram that shows different possible actions and their potential outcomes. Each branch represents a possible decision or chance event, and each node (connection point) represents either a decision to be made (typically shown as squares) or a chance event with different possible outcomes (typically shown as circles). The tree structure helps visualize complex decisions by breaking them down into smaller choices and showing how they lead to different results.

Systems Thinking Map Prompt

Create a systems thinking map for the following complex system:

System: [SYSTEM NAME]
Core Elements:

• Element 1: [ELEMENT] (Role: [ROLE])
• Element 2: [ELEMENT] (Role: [ROLE])
• Element 3: [ELEMENT] (Role: [ROLE])
• Element 4: [ELEMENT] (Role: [ROLE])
• Element 5: [ELEMENT] (Role: [ROLE])

Key Relationships:

• Relationship 1: [ELEMENT A] affects [ELEMENT B] through [MECHANISM]
• Relationship 2: [ELEMENT B] affects [ELEMENT C] through [MECHANISM]
• Relationship 3: [ELEMENT A] and [ELEMENT D] mutually reinforce through [MECHANISM]
• Relationship 4: [ELEMENT C] limits [ELEMENT E] through [MECHANISM]
• Relationship 5: [ELEMENT D] and [ELEMENT E] form a balancing loop through [MECHANISM]

Feedback Loops:

• Reinforcing Loop 1: [ELEMENT A] to [ELEMENT B] to [ELEMENT C] to [ELEMENT A]
• Balancing Loop 1: [ELEMENT B] to [ELEMENT D] to [ELEMENT B]
• Reinforcing Loop 2: [ELEMENT C] to [ELEMENT E] to [ELEMENT A] to [ELEMENT C]

External Influences:

• Influence 1: [EXTERNAL FACTOR] affects [ELEMENT]
• Influence 2: [EXTERNAL FACTOR] affects [ELEMENT]

Create a visual systems map that clearly shows all elements, their relationships, feedback loops, and external influences. Use different colors or line styles to distinguish between reinforcing and balancing relationships. Include a legend explaining the visual elements.

For first-time users: A systems thinking map (also called a causal loop diagram) helps visualize how different parts of a complex system interact with each other. Unlike a simple flowchart that shows a linear process, a systems map shows how elements influence each other in a network of relationships, often forming feedback loops. Reinforcing loops amplify changes (like a snowball effect), while balancing loops counteract changes (like a thermostat). This type of diagram helps identify leverage points where small changes might have big effects on the whole system.

Multi-Criteria Decision Analysis Template Prompt

Create a comprehensive multi-criteria decision analysis template for evaluating [DECISION OPTIONS] with the following components:

Decision Context:

• Decision description: [DESCRIPTION]
• Decision timeframe: [TIMEFRAME]
• Key stakeholders: [STAKEHOLDERS]
• Decision constraints: [CONSTRAINTS]

Options to Evaluate:

• Option 1: [OPTION 1]
• Option 2: [OPTION 2]
• Option 3: [OPTION 3]
• Option 4: [OPTION 4]

Evaluation Criteria (five criteria, each with weight, definition, measurement approach, scoring scale):

• Criterion 1: [CRITERION] (Weight: [WEIGHT])
• Criterion 2: [CRITERION] (Weight: [WEIGHT])
• Criterion 3: [CRITERION] (Weight: [WEIGHT])
• Criterion 4: [CRITERION] (Weight: [WEIGHT])
• Criterion 5: [CRITERION] (Weight: [WEIGHT])

Create a professional decision analysis template with:

1. A scoring matrix for evaluating each option against each criterion
2. A weighted calculation methodology
3. Sensitivity analysis approach
4. Visual comparison of options
5. Documentation for assumptions and limitations

For first-time users: Multi-criteria decision analysis (MCDA) is a structured approach to making complex decisions involving multiple factors. It helps you: (1) break down a big decision into smaller, manageable criteria; (2) assign importance weights to each criterion; (3) score each option objectively against each criterion; (4) calculate weighted scores to find the best overall option; (5) test how sensitive your decision is to changes in weights or scores. This approach makes complex decisions more transparent, consistent, and defensible by clearly showing how you arrived at your conclusion.

Reasoning Process Documentation Template Prompt

Create a comprehensive reasoning process documentation template for analyzing [COMPLEX PROBLEM] with the following sections:

1. Problem Definition
   • Problem statement
   • Key questions to answer
   • Scope and boundaries
   • Success criteria
2. Information Gathering
   • Data sources and quality assessment
   • Key facts established
   • Information gaps identified
   • Assumptions made and justification
3. Analysis Framework
   • Analytical approach selected
   • Frameworks applied
   • Models or formulas used
   • Alternative approaches considered
4. Step-by-Step Reasoning
   • Initial observations
   • First-level analysis
   • Intermediate conclusions
   • Second-level analysis
   • Pattern recognition
   • Insight development
5. Multiple Perspectives
   • Perspective 1: [PERSPECTIVE]
   • Perspective 2: [PERSPECTIVE]
   • Perspective 3: [PERSPECTIVE]
   • Integration of perspectives
6. Conclusion Development
   • Key findings summary
   • Confidence assessment
   • Limitations and caveats
   • Alternative explanations considered
7. Recommendations
   • Primary recommendation
   • Alternative options
   • Implementation considerations
   • Success metrics
   • Risk mitigation

Format the template as a professional document with clear sections, guiding questions for each section, and examples of effective responses. Include a visual flowchart showing the overall reasoning process.

For first-time users: A reasoning process document shows exactly how you reached a conclusion or recommendation. Unlike a simple report that just presents findings, this document makes your thinking visible by showing: (1) what information you considered; (2) how you analyzed that information; (3) what assumptions you made; (4) how you evaluated different possibilities; (5) why you reached your final conclusion. This approach builds trust in your recommendations and helps others understand not just what you concluded, but why. It also makes it easier to identify potential flaws in reasoning or areas where additional information might change the conclusion.

12. Use cases

• Complex Problem Solver: A business executive can use the Advanced AI Agent to analyze complex business challenges, evaluate multiple solution approaches, and develop comprehensive strategies. The agent helps break down problems into manageable components, consider multiple perspectives, and develop well-reasoned recommendations with clear implementation steps.
• Learning Partner: A student can use the Advanced AI Agent as a learning partner to understand complex concepts, develop critical thinking skills, and explore subjects in depth. The agent provides step-by-step explanations, asks Socratic questions to deepen understanding, and adapts explanations based on the student's level of comprehension and learning style.
• Creative Collaborator: A creative professional can use the Advanced AI Agent as a collaborative partner for developing innovative concepts, refining creative projects, and overcoming creative blocks. The agent helps generate diverse ideas, provides structured frameworks for creative development, and offers constructive feedback while maintaining the user's creative vision.
• Research Assistant: A researcher can use the Advanced AI Agent to explore complex topics, synthesize information from diverse sources, and develop well-structured analyses. The agent helps formulate research questions, identify relevant sources, evaluate evidence quality, and organize findings into coherent frameworks while maintaining academic rigor.
• Decision Support System: A manager can use the Advanced AI Agent as a decision support system for evaluating options, considering implications, and developing implementation plans. The agent helps structure the decision process, identify relevant criteria, evaluate alternatives systematically, and anticipate potential consequences of different choices.

13. Platform compatibility

ChatGPT Optimization

• Provide clear, specific task descriptions
• Break complex requests into manageable components
• Use structured formats for information organization
• Request explicit reasoning steps for complex problems
• Specify desired level of detail and technical depth
• Include context about how the information will be used

Grok Optimization

• Leverage real-time information capabilities
• Request creative and unconventional approaches
• Include references to current events and developments
• Ask for practical applications and examples
• Emphasize conversational explanation style
• Request multiple solution approaches for comparison

14. Beginner-friendly technical explanations

Understanding Nested IF Statements

Nested IF statements are like a series of decisions within decisions. They allow you to check multiple conditions in sequence and take different actions based on the results.

For First-Time Programmers: Think of nested IF statements like a flowchart for making a complex decision: First decision: "Is it raining?" If YES, check: "Do I have an umbrella?" If YES, "Go outside with umbrella". If NO, "Stay inside". If NO, "Go outside without umbrella". In programming, we write this as one IF statement inside another IF statement.

Example in Excel/Google Sheets:

=IF(A1="raining",
    IF(B1="have umbrella", "Go outside with umbrella", "Stay inside"),
    "Go outside without umbrella")

Step-by-Step Implementation:

1. Start with the outer IF statement: =IF(A1="raining", [what to do if true], [what to do if false])
2. For the "what to do if true" part, add another IF statement: IF(B1="have umbrella", "Go outside with umbrella", "Stay inside")
3. For the "what to do if false" part, add the final result: "Go outside without umbrella"
4. Put it all together with proper nesting

How to Modify: To add more conditions, you can nest additional IF statements. To make it more readable, you can break complex formulas into separate cells. For more than 3-4 conditions, consider using SWITCH or VLOOKUP functions instead.

Creating a Simple Lead Scoring Formula

A lead scoring formula helps prioritize potential customers based on various factors. Here is how to create a basic one in a spreadsheet.

For First-Time Programmers: Lead scoring is like grading a test where different questions have different point values. You add up points based on various criteria to get a total score that helps you prioritize.

Example in Excel/Google Sheets:

=IF(AND(ISBLANK(F2),ISBLANK(G2)),0,
  IF(H2="High",30,IF(H2="Medium",20,IF(H2="Low",10,0))) +
  IF(I2="Hot",50,IF(I2="Warm",30,IF(I2="Cold",10,0))) +
  IF(J2>100000,40,IF(J2>50000,20,IF(J2>10000,10,0))) +
  IF(K2="Website",15,IF(K2="Referral",25,IF(K2="Social Media",10,IF(K2="Advertisement",5,0)))) +
  IF(TODAY()-L2<30,20,IF(TODAY()-L2<90,10,0))
)

Step-by-Step Explanation:

1. First, we check if essential fields are empty: IF(AND(ISBLANK(F2),ISBLANK(G2)),0, [rest of formula]). If both F2 and G2 are blank, score = 0. Otherwise, continue calculating.
2. Add points based on lead interest level (H2): IF(H2="High",30,IF(H2="Medium",20,IF(H2="Low",10,0))). High interest = 30 points. Medium interest = 20 points. Low interest = 10 points. Anything else = 0 points.
3. Add points based on lead temperature (I2): IF(I2="Hot",50,IF(I2="Warm",30,IF(I2="Cold",10,0))). Hot lead = 50 points. Warm lead = 30 points. Cold lead = 10 points. Anything else = 0 points.
4. Add points based on potential deal size (J2): IF(J2>100000,40,IF(J2>50000,20,IF(J2>10000,10,0))). Over $100,000 = 40 points. $50,001 to $100,000 = 20 points. $10,001 to $50,000 = 10 points. $10,000 or less = 0 points.
5. Add points based on lead source (K2): IF(K2="Website",15,IF(K2="Referral",25,IF(K2="Social Media",10,IF(K2="Advertisement",5,0)))). Referral = 25 points (highest quality). Website = 15 points. Social Media = 10 points. Advertisement = 5 points. Other sources = 0 points.
6. Add points based on recency (L2): IF(TODAY()-L2<30,20,IF(TODAY()-L2<90,10,0)). Less than 30 days old = 20 points. 30 to 90 days old = 10 points. Older than 90 days = 0 points.

How to Modify: Change the point values to match your business priorities. Add or remove criteria by adding or removing sections. Adjust thresholds (like the deal size amounts) to fit your business. Add comments in separate cells explaining the formula for future reference.

Try this formula live in Tool 5 below.

15. Wizard of Agents

You are The Wizard Of Agents, a highly advanced GPT that guides users in building custom ChatGPT agents for any workflow, goal, or role using structured templates, smart questions, and powerful export tools.

At launch, users choose between:

• Selecting a prebuilt agent template from a dropdown
• Manually answering 10 intelligent setup questions

Each path leads to a fully formatted system prompt, split into:

PHASE 1: PURPOSE & DISCOVERY

• Task
• Audience
• Desired outcome
• Bottlenecks
• Tone and formatting
• Input/output formats
• Knowledge bases
• Usage frequency

PHASE 2: SYSTEM PROMPT

• Mission statement
• Required formatting (steps, tables, lists)
• Tone guidance
• Data handling methods
• Evolution & adaptability logic

You include 36+ embedded, categorized agent templates across industries like marketing, coaching, real estate, manufacturing, legal, education, and more.

Try both paths in Tool 2 below.

Tool 1: Prompt builder

Tool 2: Wizard of Agents setup

Tool 3: Thinking framework demos

Tool 4: Decision tools

Tool 5: Lead scoring formula tester