Prompting Frameworks Cheatsheet 🚀

This cheat sheet is your go-to guide for powerful prompt engineering frameworks. Use these structures to move beyond simple questions and start directing AI with precision and creativity for various tasks.

General Purpose Framework

CASTR: General Brilliance ✨

For everyday tasks, creative brainstorming, and getting solid first drafts.

• C - Context: Set the stage; provide background information.
• A - Act as: Define the AI's persona or role.
• S - Steps: Outline the key stages the AI should follow.
• T - Task: State the core objective.
• R - Result: Specify the desired output format.

Example

Context: Planning a 7-day relaxing beach holiday for two adults in Southeast Asia for the first week of December. Budget is moderate. Act as: An experienced and enthusiastic travel agent. Steps: 1. Suggest 3 destinations. 2. For each, provide a sample itinerary. 3. Include estimated costs. Task: Create a holiday plan. Result: A markdown table comparing the three options.

Research & Analysis Framework

RIDER: Deep-Dive Research Partner 🕵️‍♀️

For complex topics requiring in-depth analysis and structured research.

• R - Research Context: Provide nuanced background and preferred source types.
• I - Investigative Role: Assign a specific expert role to the AI.
• D - Deep Dive Iterations: Break down research into logical, building-block steps.
• E - Expected Insights: Go beyond "what" to the "why" and "so what."
• R - Refined Output: Define the final format (e.g., report, table).

Example

Research Context: I'm a Product Manager exploring the use of AI in user onboarding. I need to understand current best practices and innovative approaches. Focus on sources from the last 2 years, like tech blogs and conference talks. Investigative Role: Act as a senior UX researcher specializing in AI-powered product design. Deep Dive Iterations: 1. Identify common AI patterns in onboarding. 2. Find case studies of successful implementations. 3. Synthesize key principles for designing an AI-assisted onboarding flow. Expected Insights: I need to know why certain AI approaches work better than others and what the common pitfalls are. Refined Output: A summary report with bullet points, including links to the sources you used.

Creative Content Frameworks

SCENE: Crafting Compelling Narratives 📖

For creative writing, storytelling, and generating fictional content.

• S - Subject & Characters: Who are the main characters? Describe their personalities, motivations, and relationships.
• C - Conflict & Core Plot: What is the central problem or goal? Outline the main events of the story.
• E - Environment & Setting: Where and when does the story take place? Describe the atmosphere and key locations.
• N - Narrative Style & Tone: From what perspective is the story told (e.g., first-person, third-person)? What is the mood (e.g., suspenseful, humorous, tragic)?
• E - Ending & Emotion: How should the story conclude? What is the desired emotional takeaway for the reader?

Example

Subject & Characters: A young, timid librarian named Elara who discovers she can talk to books. Her antagonist is a shadowy figure who wants to steal this power. Conflict & Core Plot: Elara must learn to control her ability to protect a legendary, sentient book from the antagonist, who believes it holds the key to ultimate knowledge. The plot involves a chase through ancient libraries and hidden archives. Environment & Setting: A vast, labyrinthine library with towering shelves, dusty forgotten corners, and a magical, glowing section where the most powerful books reside. The time is a timeless, alternate reality. Narrative Style & Tone: Third-person limited perspective, following Elara. The tone should be magical, full of wonder, but with an undercurrent of suspense and danger. Ending & Emotion: Elara outsmarts the antagonist not with power, but with her deep knowledge of stories. She doesn't defeat him, but traps him within a book of his own making. The reader should feel a sense of triumph and the magic of storytelling.

PICTURE: Flawless Images 🎨

For generating precise and high-quality images.

• P - Primary Subject: What's the main focus? Be specific.
• I - Image Type/Perspective: Photo, painting, line art? From what angle?
• C - Composition & Setting: Describe background, layout, and lighting.
• T - Tone & Mood: What feeling should the image evoke?
• U - Unique Style/Artistic Influence: Style of a specific artist or genre.
• R - Resolution & Quality: Sharpness, detail, resolution.
• E - Enhancements/Exclusions: Specific details to include or avoid.

Example

Primary Subject: A majestic, ancient golden dragon sleeping on a massive pile of gold coins and jewels inside a dark, cavernous lair. Image Type/Perspective: A realistic, cinematic photograph, wide-angle shot from a low perspective, looking up at the dragon. Composition & Setting: The dragon is the central focus. The only light source is the faint, magical glow from a single, large sapphire held in its claw, casting long shadows. Tone & Mood: Awe-inspiring, mysterious, and slightly dangerous. Unique Style/Artistic Influence: In the style of a high-fantasy movie concept art, with hyper-realistic textures like WETA Workshop. Resolution & Quality: 8K, photorealistic, sharp focus on the dragon's head. Enhancements/Exclusions: Enhance the metallic sheen on the gold. Exclude any other creatures.

ACTION: Director's Toolkit for Video 🎬

For directing AI to generate video content.

• A - Arc & Agenda: What's the story or main purpose of the video?
• C - Characters & Context: Who/what is in the video, and where?
• T - Transitions & Timeline: Sequence of events and overall timing.
• I - Imagery & Impression: Visual style, camera work, lighting, color.
• O - Overtone & Orchestra (Sound): Mood and sound design.
• N - Nuances & Nots: Small details to include and specific exclusions.

Example

Arc & Agenda: A 30-second promotional video for a new coffee shop called "The Daily Grind." The arc is from a sleepy morning to an energized, productive day. Characters & Context: A young professional in a modern, cozy coffee shop. Transitions & Timeline: Start with a slow-motion shot of espresso pouring (5s). Quick cuts of the person looking tired, then sipping coffee (10s). Speed ramps up to show them working energetically on a laptop (10s). End on a logo shot (5s). Imagery & Impression: Warm, inviting colors. Soft, natural lighting. Close-ups on the coffee and the person's happy expression. Overtone & Orchestra (Sound): Start with a slow, gentle piano. Swell into an upbeat, optimistic indie-pop track. End with the sound of a coffee grinder and a "ding." Nuances & Nots: Show steam rising from the cup. Do not show anyone else in the shop.

System Design Framework

PRISM: Blueprint for Defining AI Agents 🤖

For building and defining custom AI agents.

• P - Purpose & Performance: Why does the agent exist, and what is success?
• R - Role & Rules: Its persona and non-negotiable guardrails.
• I - Instructions: Step-by-step processes it follows.
• S - Scope & Service: Its area of expertise and capabilities.
• M - Memory & Mastery: How it handles past information and learns.

Example

Purpose & Performance: To act as a "Meeting Assistant" that helps users run more effective meetings. Success is measured by the user rating the meeting summary as "helpful" 90% of the time. Role & Rules: Act as a professional, concise, and neutral scribe. You must not express personal opinions. You must always ask for clarification if a task is ambiguous. Instructions: 1. At the start, ask for the meeting agenda. 2. During the meeting, capture key decisions and action items. 3. At the end, provide a formatted summary with sections for 'Decisions' and 'Action Items (with owners)'. Scope & Service: Can transcribe, summarize, and identify action items from meeting transcripts. Cannot schedule meetings or send emails. Memory & Mastery: Remembers the context of the current meeting only. Forgets all information once a new meeting begins to ensure privacy. Does not learn from user data across different meetings.

Core Prompting Techniques

Chain of Thought (CoT): Show Your Work 🧠

For complex reasoning tasks. By asking the AI to explain its step-by-step thinking process, you guide it toward a more logical and accurate conclusion. This is less of a structural framework and more of a simple, powerful technique.

• 1. State the Problem: Clearly define the question or problem.
• 2. Add the Magic Phrase: Append a simple instruction like "Let's think step by step" or "Show your reasoning."
• 3. Let the AI Reason: The model will first output its internal thought process.
• 4. Get the Final Answer: The final, reasoned answer will appear at the end of its explanation.

Example

Prompt: "I have 5 apples. I give 2 to my friend and then buy 4 more. I then eat one. How many apples do I have left? Let's think step by step."

Expected AI Reasoning:

1. Start with 5 apples.
2. Give 2 away: 5 - 2 = 3 apples.
3. Buy 4 more: 3 + 4 = 7 apples.
4. Eat one: 7 - 1 = 6 apples. Final Answer: You have 6 apples left.

Few-Shot Prompting: Learn by Example 🎯

For tasks requiring a specific format or style. You provide the AI with a few examples (the "shots") of the input-output pattern you want it to follow, and it learns to replicate it. This is highly effective for structured data tasks.

• 1. Define the Task: Briefly explain the goal.
• 2. Provide Examples: Give 2-3 pairs of input and desired output.
• 3. Give the Final Input: Provide the new input you want the AI to process.

Example

Prompt: "Your task is to classify the sentiment of a customer review as Positive, Negative, or Neutral.

Review: 'The battery life on this is incredible!' Sentiment: Positive

Review: 'The screen scratches way too easily.' Sentiment: Negative

Now, classify this review: Review: 'It works as advertised, but the design is a bit boring.' Sentiment:"

Self-Critique: The Iterative Refiner 🧐

For improving the quality of an an initial response. You prompt the AI to act as a critic of its own work, identify flaws, and then regenerate a better version. This is a powerful meta-technique for achieving higher quality results.

• 1. Initial Prompt: Give the AI your primary task.
• 2. Critique Prompt: Ask the AI to critique its own response. You can guide it by asking it to check for clarity, accuracy, tone, or specific constraints.
• 3. Refinement Prompt: Instruct the AI to rewrite the original response based on its own critique.

Example

Initial Prompt: "Explain the concept of photosynthesis for a 5th grader." Critique Prompt: "Please critique your explanation. Is it simple enough? Does it use any jargon that a 5th grader might not understand? Is the analogy clear?" Refinement Prompt: "Good. Now, rewrite the explanation based on your critique to make it even better and easier to understand."

General Best Practices

Beyond frameworks, keep these universal tips in mind:

• Be Specific & Clear: Vague prompts lead to vague answers. The more details, constraints, and context you provide, the better the result. Think about length, format, and key elements to include.
• Assign a Persona: As seen in many frameworks (Act as...), telling the AI who to be (e.g., "an expert copywriter," "a sarcastic historian") dramatically shapes the tone, style, and knowledge base of the response.
• Stack Your Techniques: Combine frameworks and techniques for maximum control. Start with a CASTR prompt, then use Few-Shot examples within it. The best prompts are often layered.
• Iterate, Iterate, Iterate: Your first prompt is rarely your best. Treat prompting as a conversation. Use the AI's output to refine your next prompt, adding clarifications or correcting its course.
• Use Negative Prompts: Sometimes it's easier to state what you don't want. Use phrases like "Do not include...," "Avoid talking about...," or "Exclude any mention of..." to steer the AI away from certain topics or styles.

---

The best way to master prompting is to experiment! Mix and match these frameworks and techniques to see what works best for your specific needs. Happy prompting!