Zero-Shot Prompting
Zero-shot prompting asks the model to complete a task directly without examples. It relies on the model's pretrained knowledge and works best when the task is simple, clear, and familiar.
Prompt Techniques
Fifteen core techniques covering basic prompting, reasoning, behavior control, architecture, and quality assurance.
Prompt Techniques
Few-Shot Prompting
Few-shot prompting teaches the model a pattern by showing a few input-output examples before the target task. It is one of the most reliable ways to improve format and style consistency.
Chain-of-Thought (CoT)
Chain-of-thought prompting asks the model to reason step by step before giving an answer. It is useful for multi-step reasoning, calculations, and transparent decision workflows.
Tree-of-Thought (ToT)
Tree-of-thought extends step-by-step reasoning by exploring multiple candidate paths, scoring them, and selecting the strongest solution.
System Prompt Design
System prompt design defines the assistant's role, responsibilities, boundaries, and output behavior. It is the behavior contract for an AI application.
Role-Playing Prompting
Role-playing prompting gives the model a professional identity, perspective, and communication style so it can respond with domain-specific framing.
Output Formatting
Output formatting tells the model exactly what structure to return, such as JSON, tables, markdown sections, or XML-like blocks.
Temperature Control
Temperature controls randomness. Lower values produce more deterministic outputs; higher values produce more variation and creative alternatives.
Context Window Management
Context window management decides what information enters the model input and in what priority order. It is critical for long documents, chat history, and RAG systems.
Tool Use Prompting
Tool use prompting defines when and how the model should call external tools such as search, calculators, APIs, databases, or internal functions.
Multi-Turn Strategy
Multi-turn strategy manages state, context, clarification, and topic shifts across a conversation.
Prompt Chaining
Prompt chaining splits a complex task into smaller prompts where one step's output becomes the next step's input.
Self-Consistency
Self-consistency generates multiple reasoning paths and aggregates the final answer by voting or confidence comparison.
Retrieval-Augmented Generation
Retrieval-augmented generation inserts relevant external documents into the prompt so answers are grounded in current or private knowledge.
Prompt Evaluation Methods
Prompt evaluation turns prompt tuning from trial-and-error into an engineering workflow with test sets, metrics, automated checks, and version history.