RIPE Framework
A structured approach to writing effective AI prompts using four components: Role, Instruction, Parameter, and Example.
What is the RIPE Framework?
The RIPE Framework is a structured approach to writing effective AI prompts and instructions. RIPE stands for:
- Role - Who the AI should be
- Instruction - What to do
- Parameter - Constraints and specifications
- Example - Demonstrations of desired output
This simple framework helps anyone—from beginners to experts—create prompts that get better, more consistent results from AI agents and large language models.
Why RIPE works
Reduces ambiguity Each component addresses a different aspect of the task, leaving less room for misinterpretation.
Improves consistency The same RIPE prompt produces similar results across sessions, unlike vague requests.
Easy to remember Four components, one acronym. Even non-technical users can apply it immediately.
Scales to complexity Works for simple questions and complex agentic workflows alike.
The four components
Role Define who the AI should be. This activates relevant knowledge and communication patterns.
- "You are a senior AI researcher writing for a technical audience"
- "You are a patient teacher explaining concepts to a beginner"
Instruction State clearly what to do. Be specific about the action and outcome.
- "Write a follow-up email to a prospect who attended our demo but hasn't responded"
Parameter Set constraints: length, format, tone, audience, exclusions, inclusions.
- Length: "In 200 words or less"
- Format: "As a bullet list"
- Tone: "Professional but friendly"
Example Show what good output looks like with input/output pairs. Examples often communicate better than descriptions.
RIPE in action
Simple example: Email writing
Role: Professional business writer helping with client communication Instruction: Write a thank-you email to a new client Parameters: 3-4 sentences, warm but professional, include next steps Example: [Previous successful email template]
Complex example: Agent configuration Use RIPE in SOUL.md and CLAUDE.md files:
- Role section defines agent identity
- Instruction section lists behavioral guidelines
- Parameter section sets constraints and boundaries
- Example section shows sample interactions
Common mistakes
Weak roles ❌ "You are helpful" ✅ "You are a senior data analyst who explains findings to non-technical executives"
Vague instructions ❌ "Write something about marketing" ✅ "Create a marketing email announcing our new feature to existing customers"
Missing parameters ❌ "Make it good" ✅ "Under 200 words, professional, include clear call to action"
Unhelpful examples ❌ "Like my other emails" ✅ [Actual example of a successful email]
The RIPE Framework is a component of effective context engineering—use it within your system prompts, agent configurations, and individual requests.
Related Terms
Prompt Engineering
The practice of designing and refining inputs to AI models to get better, more accurate, and more useful outputs.
Context Engineering
The practice of designing and optimizing all information provided to an AI model—including system prompts, retrieved documents, conversation history, tools, and examples—to get better results.
System Prompt
Special instructions given to an AI model that define its behavior, personality, and constraints before any user interaction.
Few-Shot Learning
Teaching AI models to perform tasks by providing a small number of examples (typically 1-10) in the prompt.
Apply RIPE to your AI agents
Build AI agents on Chipp with templates that follow proven prompting frameworks.
Learn more