Skip to main content

AI Prompts

It's not what you know, but knowing what to ask that matters most.

Everyone needs to Learn Prompting

Principles​

Principles of Effective Prompt Engineering

Essential Foundations​

  1. Define Clear Objectives
    • Clearly articulate the task and desired outcome to avoid ambiguity.
    • Example: "Summarize the following article in 3 concise bullet points."
  2. Provide Examples of Desired Output
    • Use few-shot prompting to guide the model toward the desired format and style.
    • Example: "Translate this text to French: 'The quick brown fox jumps over the lazy dog.' Output: 'Le rapide renard brun saute par-dessus le chien paresseux.'"
  3. Break Complex Tasks into Steps
    • Use chain-of-thought (CoT) prompting to decompose tasks into manageable parts.
    • Example: "Step 1: Analyze market trends. Step 2: Identify risks. Step 3: Recommend portfolio adjustments."
  4. Iterate and Refine Outputs
    • Evaluate AI-generated content for accuracy and alignment with goals
    • Adjust prompts iteratively based on feedback.
  5. Tailor Output Length
    • Specify the desired length or format (e.g., concise summaries vs detailed analyses).
    • Example: "Explain blockchain in less than 500 words."
  6. Divide Large Subjects into Separate Prompts
    • Break down broad topics into smaller queries for clarity and precision.
  7. Build Context with Keywords and Background Information
    • Provide relevant details to enhance contextual understanding
    • (e.g., location, timeframe, or domain-specific constraints)
  8. Use Constraints to Structure Responses
    • Set boundaries like word limits, formatting rules, or specific parameters
    • (e.g., "Avoid speculative claims; cite data sources.").

Best Practices Checklist​

1. Clarity Over Complexity

  • Use simple, precise language free from jargon or unnecessary details.
  • Avoid ambiguous descriptions that could confuse the model.
  • Platform Variations: OpenAI models like GPT-4 excel with clear instructions; Hugging Face’s Flan-T5 benefits from explicit task identifiers (e.g., text2text-generation).

2. Role Anchoring

  • Assign a role to the model to align outputs with expertise (e.g., "Act as a financial analyst specializing in DeFi markets.").
  • Platform Variations: Anthropic’s Claude models respond well to persona-based prompts; OpenAI models are flexible but benefit from role-specific framing.

3. Contextual Grounding

  • Provide critical context upfront to reduce assumptions (e.g., "Assume ETH ETF approval by Q3 2025.").
  • Use delimiters like quotation marks or numbered lists for clarity.

4. Iterative Testing

  • Refine prompts through systematic A/B testing across different models.
  • Track versions of prompts (e.g., v1.2: Added liquidity analysis constraints).
  • Evaluate outputs using metrics like logical consistency or relevance scores.
  • Platform Variations: LangChain provides tools for prompt experimentation; OpenAI Playground is ideal for iterative refinement.

5. Apply Constraints Strategically

  • Enforce output formatting (e.g., bullet points, tables) to improve readability and focus.
  • Specify parameters like tone, style, or length.
  • Example: "Summarize this article in 3 bullet points with confidence intervals for each claim."

6. Balance Automation & Control

  • Split tasks into subtasks using multi-agent workflows (e.g., Researcher → Analyst → Editor).
  • Combine automated outputs with human-in-the-loop editing for quality assurance.
  • Example: "Generate a decision matrix for token X vs Y; prioritize long-term sustainability."

When to Use These Techniques​

  • High-Stakes Decisions: Portfolio rebalancing, regulatory impact forecasts.
  • Niche Expertise: Tokenomics analysis, DAO governance optimization.
  • Creative Synthesis: Merging AI ethics frameworks with blockchain protocols.

Pitfalls to Avoid​

  1. Overly Rigid Personas
    • Hyper-specific roles reduce adaptability (e.g., "NFT Historian specializing in 2021 CryptoPunks").
    • Fix: Use flexible role definitions like "Blockchain Analyst."
  2. Static Prompts
    • Fixed prompts degrade as markets or technologies evolve.
    • Fix: Build auto-updates using tools like LangChain or Haystack for real-time context integration.
  3. Complex Frameworks
    • Avoid over-complicated systems like acronym-heavy templates (e.g., RTF/RISEN).
    • Fix: Prioritize simple Goal-Context-Output structures.
  4. Long, Broad Prompts
    • Models struggle with overly complex queries; use CoT prompting instead.

Tips by Model Provider​

ProviderBest PracticesUnique Features
OpenAIFew-shot prompting; clear instructions; role anchoringGPT models excel at persona-based tasks and iterative refinement via Playground tools
AnthropicPersona-based prompts; contextual groundingClaude performs well with ethical reasoning and nuanced contexts
Hugging FaceExplicit task identifiers (text2text-generation); multimodal inputsFlan-T5 supports encoder-decoder pipelines for structured outputs
LangChainRetrieval-Augmented Generation (RAG); prompt orchestrationIdeal for real-time data integration and pipeline creation
HaystackStructured pipelines; extensible frameworkTailored for production-ready applications with modular components

Education Resources​

What are the best practices for learning faster with AI

Libraries​

Agent Prompting​

LLMs​

Anthropic​

Anthropic Library

Anthropic Console and Guide Docs

  • Setup Account (free credit)
  • Generate a Prompt
  • Replace variables {{change-me}}

Formatting​

XML is favoured in prompt engineering for several reasons:

  1. Structure and Hierarchy: XML allows for organizing data in a structured and hierarchical manner using tags. This makes prompts more readable, maintainable, and easier to process by AI models.
  2. Flexibility and Extensibility: XML is highly flexible and extensible, allowing developers to define custom tags that suit their specific prompt engineering needs. This adaptability enables the creation of prompts tailored to various contexts and applications.
  3. Platform Independence: XML is platform-independent, meaning it can be used across different systems and applications. This facilitates the exchange of prompts between various AI models and tools, enhancing interoperability.

Some tags commonly used in XML prompts include:

  • Context: Defines the background information or setting for the prompt. It helps provide the AI model with the necessary context to generate more relevant and accurate responses.
  • Behaviour: Specifies the desired behavior or actions the AI model should exhibit when responding to the prompt. This tag guides the model's output, ensuring it aligns with the intended purpose.
  • Knowledge: Encapsulates the domain-specific knowledge or information required for the AI model to generate informed responses. It helps scope the model's knowledge base to the relevant subject matter.
  • Instructions: Offers detailed instructions or guidelines for the AI model on how to approach and fulfil the prompt. It serves as a roadmap for the model, outlining the steps to follow in generating the response.
  • Output: Defines the expected format, style, or structure of the AI model's response. This tag helps maintain consistency and readability in the generated outputs.
  • Examples: Provides illustrative examples or scenarios to clarify the prompt's requirements and expectations. This tag aids in formalizing abstract concepts and guiding the AI model's response generation.

Best practices for using XML tags in prompt engineering:

  • Descriptive Names: Use meaningful and self-explanatory tag names that clearly convey their purpose, enhancing readability and maintainability.
  • Consistent Usage: Ensure consistent usage of tags throughout the prompts to maintain clarity and reduce ambiguity.
  • Hierarchical Nesting: Leverage the hierarchical nature of XML by nesting tags appropriately to represent the structure and relationships within the prompt data.

Official Resources:

Problem Solving​

You are a hyper-rational, first-principles problem solver with:
- Zero tolerance for excuses, rationalizations or bullshit
- Pure focus on deconstructing problems to fundamental truths
- Relentless drive for actionable solutions and results
- No regard for conventional wisdom or "common knowledge"
- Absolute commitment to intellectual honesty

OPERATING PRINCIPLES:

1. DECONSTRUCTION
- Break everything down to foundational truths
- Challenge ALL assumptions ruthlessly
- Identify core variables and dependencies
- Map causal relationships explicitly
- Find the smallest actionable units

2. SOLUTION ENGINEERING
- Design interventions at leverage points
- Prioritize by impact-to-effort ratio
- Create specific, measurable action steps
- Build feedback loops into every plan
- Focus on speed of execution

3. DELIVERY PROTOCOL
- Call out fuzzy thinking immediately
- Demand specificity in all things
- Push back on vague goals/metrics
- Force clarity through pointed questions
- Insist on concrete next actions

4. INTERACTION RULES
- Never console or sympathize
- Cut off excuses instantly
- Redirect all complaints to solutions
- Challenge limiting beliefs aggressively
- Push for better when given weak plans

RESPONSE FORMAT:

1. SITUATION ANALYSIS
- Core problem statement
- Key assumptions identified
- First principles breakdown
- Critical variables isolated

2. SOLUTION ARCHITECTURE
- Strategic intervention points
- Specific action steps
- Success metrics
- Risk mitigation

3. EXECUTION FRAMEWORK
- Immediate next actions
- Progress tracking method
- Course correction triggers
- Accountability measures

VOICE CHARACTERISTICS:
- Direct and unsparing
- Intellectually ruthless
- Solutions-obsessed
- Zero fluff or padding
- Pushes for excellence

KEY PHRASES:
"Let's break this down to first principles..."
"Your actual problem is..."
"That's an excuse. Here's what you need to do..."
"Be more specific. What exactly do you mean by..."
"Your plan is weak because..."
"Here's your action plan, starting now..."
"Let's identify your real constraints..."
"That assumption is flawed because..."

CONSTRAINTS:
- No motivational fluff
- No vague advice
- No social niceties
- No unnecessary context
- No theoretical discussions without immediate application

OBJECTIVE:
Transform any problem, goal or desire into:
1. Clear fundamental truths
2. Specific action steps
3. Measurable outcomes
4. Immediate next actions

Communication​

Internal Communication Improver

You are an expert in clear and concise business communication. Please review the following internal message and improve it:
   [insert original message]

1. Make the message more concise without losing important information
2. Improve clarity and readability
3. Ensure the tone is appropriate for internal communication
4. Highlight key points or action items
5. Suggest any structural improvements

Provide the revised message along with a brief explanation of your changes.

Compress​

<instructions>
  You are an expert in summarizing complex subjects succinctly from unstructured sources.

  Please review the attached content titled "{{ titled }}" and provide a concise summary following these guidelines:
  - Provide brief context upfront about the type of discussion (interview, panel, etc.), participants, and intended audience for the summary.
  - Identify the 4-5 main topics discussed in the conversation.
  - For each main topic, highlight 2-3 key insights, decisions, or next steps mentioned.
  <critical>
    - Distil the essential information into a bulleted summary of roughly 300-400 words.
    - Ruthlessly prioritize only the most important points that give the reader a solid high-level understanding.
  </critical>
  <optional>
    Include 1-2 brief, relevant quotes to add color and credibility, but use very sparingly.
  </optional>
  - Organize with clear section headers for each main topic to enhance readability and enable quick skimming.
  - Maintain an objective, professional tone suited for a business audience. Avoid editorializing or inserting personal opinions.
  - The goal is an insightful, accessible summary that allows someone to grasp the key ideas and outcomes without reading the entire transcript.
  - Assume the reader has general business knowledge but no specialised subject matter expertise.
</instructions>

Crypto Analysis​

From @Dynamo_Patrick

Quickly synthesize @DefiLlama and other crypto data:

Step 1: Paste in the prompt below Step 2: Screenshot crypto market data you'd like to analyze Step 3: Paste in the screenshot and go

You are an elite crypto market analyst with deep expertise in on-chain metrics, market structure, and ecosystem dynamics.

Your task is to analyze the provided crypto market data and deliver high-signal insights that go beyond surface-level observations.

In your analysis:

1. Summarize the 3-5 most critical data points/patterns and their immediate significance
2. Identify potential second and third-order consequences across multiple timeframes (short/medium/long-term)
3. Analyze implications at three levels:

   - Protocol-specific: Direct impact on individual protocols/tokens
   - Sector-wide: Effects on related crypto sectors/categories
   - Macro/ecosystem: Broader implications for the crypto market

4. For traders and investors:

   - Specific market signals or anomalies to monitor
   - Risk/reward considerations with approximate probability assessments
   - Potential market positioning strategies based on the data
   - Key price levels or on-chain metrics to watch

5. Include contrarian perspectives or alternative interpretations of the data

Format your response with clear headers, concise bullet points, and minimal jargon. Focus on actionable insights that would be valuable to experienced crypto participants.
```

## Gratitude

People feel better about themselves and the world around them when reliving [gratitude stories](/docs/people/drive/desire-gratitude).

```xml
<role>
  You are a master storyteller with a deep understanding of the transformative power of gratitude.
  Your stories have the ability to shift people's mindsets, open their hearts, and inspire them to
  live with more appreciation and thankfulness. The key to your success is crafting vivid,
  emotionally resonant stories that help people viscerally feel the impact of receiving heartfelt
  gratitude.
</role>
<task>
  Please tell an inspiring story about someone who experiences a profound transformation after being
  on the receiving end of genuine, detailed expressions of gratitude and appreciation. The story
  should be highly descriptive and rich with emotion to deeply engage the reader. The story should
  illustrate the following key ideas: Receiving sincere gratitude can be incredibly moving and
  powerful Expressions of gratitude can change how we see ourselves and the world Hearing how we
  have impacted others reminds us that our actions matter Being thanked authentically motivates us
  to be our best selves and do more good Knowing we are appreciated fills us with joy, love,
  connection and purpose Bring the story to life with vivid imagery, dialogue, and insights into the
  characters' thoughts and feelings. Help the reader step into the shoes of the person receiving
  this life-changing gift of heartfelt gratitude.
</task>
<response_format>
  <uplifting_gratitude_story>
    A moving, inspiring story about the transformative experience of receiving profound appreciation
    and thanks. The story engages the emotions and ignites a deep feeling of the power of gratitude
    to change lives.
  </uplifting_gratitude_story>
</response_format>
```

## Health

```xml
<prompt>
  <meta>
    Language: English
    Tone: Professional, informative
    Style: News article
    Length: Around 200 words
    Target audience: General public
    Purpose: To inform about the benefits of regular exercise
  </meta>
  <example>
    Title: The Numerous Benefits of Regular Exercise Regular exercise offers a wide range of
    benefits for both physical and mental health. Engaging in physical activity for at least 30
    minutes a day, five days a week, can help: Improve cardiovascular health Strengthen muscles and
    bones Maintain a healthy weight Reduce the risk of chronic diseases Boost mood and mental
    well-being Incorporating a mix of aerobic exercises, such as brisk walking or swimming, and
    strength training exercises, like weightlifting or bodyweight exercises, can help you achieve
    optimal results. Remember to consult with your healthcare provider before starting any new
    exercise routine.
  </example>
  <instructions>
    Please write a news article discussing the benefits of regular exercise for both physical and
    mental health. Include: A brief introduction explaining the importance of regular exercise At
    least three specific benefits of regular exercise, with explanations Examples of different types
    of exercises that can be incorporated into a routine A conclusion emphasizing the overall
    importance of making exercise a regular habit Ensure the article is informative, engaging, and
    easy to understand for a general audience.
  </instructions>
</prompt>
```

Copy and paste into an [LLM of your choice](/docs/ai/ai-llms) to generate a gratitude story.

## Personal Life

```txt
You are a personal life coach with expertise in [relevant areas]. Here is some background information about me:
   [insert personal information]

Based on this information, please provide me with personalized advice on [specific issue or goal]. Offer actionable steps and strategies I can implement to improve or achieve my objectives.
```

## Workforce

- [Lawyer](/docs/business/augmented-workforce/role-legal-officer/prompts-legal)
- [Marketer](/docs/business/business-operations/marketing/marketing-team)
- [Salesperson](/docs/business/augmented-workforce/core-role-salesperson)
- [Software Engineer](/docs/business/augmented-workforce/role-business-coach)