Prompt engineering techniques, alphabetical list

Generated By

Claude 3.5 Sonnet

Platform

API

Generation Date

Nov 14th 2024

| Analogical Reasoning | Prompts the AI to draw comparisons between different concepts or situations | Comparative Analysis | Problem-solving, Creative thinking | | Audience Specification | Defines a specific audience for the AI to address in its response | Target Audience | Tailoring content, Improving relevance | | Chain of Thought | Breaks down complex tasks into smaller steps | COT | Reasoning prompts, Problem-solving | | Character Roleplay | Instructs the AI to respond as a specific character or persona | Persona Adoption | Creative writing, Dialogue generation | | Chunking | Divides large amounts of information into smaller, manageable chunks | Information Segmentation | Data processing, Memory management | | Clarification Requests | Encourages the AI to ask for more information when needed | Follow-up Questions | Improving accuracy, Handling ambiguity | | Comparative Analysis | Asks the AI to compare and contrast different elements | Juxtaposition | Decision-making, Critical thinking | | Conditional Statements | Uses if-then statements to guide the AI's response | Logical Branching | Decision trees, Scenario planning | | Constraint Specification | Sets specific limitations or requirements for the AI's output | Boundary Setting | Focused responses, Adherence to guidelines | | Context Injection | Provides additional background information to guide the AI's understanding | Background Setting | Improving relevance, Enhancing accuracy | | Contradictory Prompting | Presents conflicting information to stimulate critical thinking | Paradoxical Prompting | Debate preparation, Critical analysis | | Counterfactual Thinking | Asks the AI to consider alternative scenarios or outcomes | What-If Analysis | Strategic planning, Creative problem-solving | | Data Formatting | Specifies the desired format for data in the AI's response | Output Structuring | Data visualization, Report generation | | Emotional Prompting | Instructs the AI to incorporate specific emotions in its response | Sentiment Injection | Creative writing, Empathy training | | Ethical Considerations | Prompts the AI to consider moral implications in its responses | Moral Reasoning | Decision-making, Policy development | | Example-Driven Prompting | Provides specific examples to guide the AI's output | Template-Based Prompting | Consistent formatting, Style mimicking | | Explicit Instruction | Gives clear, step-by-step directions for the AI to follow | Procedural Prompting | Task completion, Process explanation | | Few-Shot Learning | Provides a few examples to help the AI understand the task | Example-Based Learning | Quick task adaptation, Pattern recognition | | Fictional Scenario Creation | Asks the AI to generate imaginary situations or stories | Creative Worldbuilding | Storytelling, Hypothetical analysis | | Goal-Oriented Prompting | Clearly states the desired outcome or objective | Outcome-Focused Prompting | Problem-solving, Strategic planning | | Gradual Revelation | Reveals information incrementally to guide the AI's thought process | Progressive Disclosure | Storytelling, Mystery solving | | Historical Context | Provides relevant historical information to inform the AI's response | Temporal Framing | Analysis of trends, Historical writing | | Hypothetical Questioning | Poses "what if" scenarios to explore potential outcomes | Speculative Inquiry | Strategic planning, Risk assessment | | Iterative Refinement | Gradually refines the prompt based on previous responses | Feedback Loop | Improving accuracy, Fine-tuning output | | Knowledge Integration | Prompts the AI to combine information from multiple sources | Cross-Domain Synthesis | Interdisciplinary analysis, Comprehensive understanding | | Language Style Specification | Defines a particular language style or tone for the AI to use | Tone Setting | Content adaptation, Brand consistency | | Logical Deduction | Guides the AI through a series of logical steps to reach a conclusion | Reasoning Chain | Problem-solving, Analytical thinking | | Meta-Cognitive Prompting | Asks the AI to explain its thought process or reasoning | Self-Reflection | Transparency, Debugging AI responses | | Multi-Modal Prompting | Combines different types of input (text, images, etc.) in the prompt | Mixed Media Prompting | Rich content creation, Comprehensive analysis | | Negative Prompting | Specifies what the AI should not include in its response | Exclusion Criteria | Focused outputs, Avoiding unwanted content | | Open-Ended Questioning | Uses broad, non-specific questions to encourage exploration | Exploratory Prompting | Brainstorming, Discovering new ideas | | Perspective Shifting | Asks the AI to consider a topic from different viewpoints | Multi-Angle Analysis | Balanced analysis, Empathy development | | Persona-Based Prompting | Defines a specific persona for the AI to adopt in its responses | Character Assumption | Roleplaying, Diverse perspective generation | | Pre-Processing | Formats or structures the input data before presenting it to the AI | Input Optimization | Improving AI understanding, Enhancing efficiency | | Prioritization | Asks the AI to rank or prioritize items based on specific criteria | Hierarchical Ordering | Decision-making, Task management | | Problem Reframing | Encourages the AI to look at a problem from a different angle | Perspective Shifting | Creative problem-solving, Overcoming mental blocks | | Process Simulation | Guides the AI through a step-by-step simulation of a process | Procedural Walkthrough | Training, Process optimization | | Prompt Chaining | Links multiple prompts together in a sequence | Sequential Prompting | Complex task completion, Narrative development | | Prompt Templating | Uses pre-defined templates to structure prompts consistently | Standardized Prompting | Consistency, Efficiency in prompt creation | | Quantitative Constraints | Specifies numerical limits or requirements in the prompt | Numerical Bounding | Data analysis, Resource allocation | | Question Decomposition | Breaks down complex questions into simpler sub-questions | Query Segmentation | Handling complex queries, Improving comprehension | | Recursive Prompting | Uses the output of one prompt as input for another | Iterative Questioning | Deep exploration, Refining ideas | | Reflective Questioning | Asks the AI to evaluate or reflect on its own responses | Self-Assessment | Quality improvement, Bias detection | | Role Assignment | Assigns a specific role or job title to the AI | Professional Persona | Specialized responses, Expert simulation | | Scenario Planning | Presents different scenarios for the AI to analyze or respond to | Situational Analysis | Strategic planning, Risk assessment | | Semantic Priming | Provides context or related concepts to influence the AI's associations | Conceptual Framing | Guiding thought direction, Enhancing relevance | | Socratic Questioning | Uses a series of probing questions to stimulate critical thinking | Inquiry-Based Prompting | Deep analysis, Uncovering assumptions | | Specificity Gradients | Gradually increases the level of detail in prompts | Progressive Detailing | Focused exploration, Depth of analysis | | Task Decomposition | Breaks down complex tasks into smaller, manageable sub-tasks | Modular Prompting | Project management, Simplifying complex problems | | Time-Based Prompting | Incorporates time elements or constraints in the prompt | Temporal Framing | Scheduling, Historical or futuristic analysis | | Zero-Shot Learning | Asks the AI to perform a task without prior examples | Context-Free Prompting | Testing AI capabilities, Novel task execution | The above text was generated by a large language model (LLM) and its accuracy has not been validated. This page is part of 'LLMs-on-LLMs,' a Github repository by Daniel Rosehill which explores how curious humans can use LLMs to better their understanding of LLMs and AI. However, the information should not be regarded as authoritative and given the fast pace of evolution in LLM technology will eventually become deprecated. This footer was added at 16-Nov-2024.