In partnership with

Reading time: 6 minutes / Become my affiliate / Sponsor this newsletter

Greetings from above,

Why did the prompt engineer get fired? He kept asking the same question 47 times. (Turns out there was a better way.)

ALEX'S STORY: Three months ago, I was writing the same prompts over and over. Every conversation started with 15 minutes of context-setting. "I run a prompt engineering business. My audience is small business owners. My tone is direct and witty." Again. And again. And again.

It felt like Groundhog Day for AI. I was treating Claude like a stranger every single time we talked. The worst part? The outputs were inconsistent because I was describing my context slightly differently each session.

Then I discovered a feature most people don't know exists. It's called Skills. And it completely changed how I interact with AI.

Now I have systems that run entire decision frameworks, generate content in my exact voice, and guide me through complex problems step by step. Claude knows who I am before I say a word.

Today's workflow system will show you:

  • What Skills actually are (and why 90% of people get it wrong)

  • The built-in shortcut that creates Skills for you automatically

  • How to make Claude research your business and remember it forever

Let's build your competitive advantage!

Help us make better ads

Did you recently see an ad for beehiiv in a newsletter? We’re running a short brand lift survey to understand what’s actually breaking through (and what’s not).

It takes about 20 seconds, the questions are super easy, and your feedback directly helps us improve how we show up in the newsletters you read and love.

If you’ve got a few moments, we’d really appreciate your insight.

🎯 THE CLAUDE SKILLS SYSTEM

Everyone thinks Skills are just fancy system prompts. Wrong.

Skills are portable instruction packages that teach Claude your specific workflows. Think of them as onboarding documents for an AI. You write the instructions once. Claude loads them automatically whenever they're relevant.

Check out the official Claude Skills guide by Anthropic.

Here's the mental model that changed everything for me: if you've typed the same prompt more than five times, you should have built a Skill by now.

This workflow creates:

  • Permanent Context: Claude remembers your business, voice, and preferences across every conversation

  • Composable Systems: Stack multiple Skills together for complex workflows

  • Interactive Frameworks: Skills that ask questions back, guiding you through multi-phase processes

🔗 WORKFLOW OVERVIEW

Here's the complete system we're building today:

Step 1: Enable Skills → Access the hidden capability

Step 2: Use the Skill-Creator → Let Claude build your first Skill automatically

Step 3: Create a Context Profile → Claude researches your business and remembers forever

Each step builds on the previous. No gaps. No guesswork.

⚙️ STEP 1: ENABLE CLAUDE SKILLS

💡 What this does: Unlocks Claude's ability to load custom instruction packages. Most users don't even know this exists.

  1. Navigate to Settings > Capabilities

  2. Ensure "Code execution and file creation" is enabled

  3. Scroll to the Skills section

  4. Toggle individual skills on or off as needed

  5. To add custom skills, click "Upload skill" and upload a ZIP file containing your skill folder

⚙️ STEP 2: THE SKILL-CREATOR SHORTCUT

💡 What this does: Claude has a built-in skill called "skill-creator" that builds skills for you. You don't need to understand file structures. You don't need to write YAML. You just describe what you want.

Important: Don't assume Claude will automatically load the skill. Be explicit.

Type this exactly: "Use the skill-creator skill to create a skill for [your workflow]"

Examples:

  • "Use the skill-creator skill to create a skill for writing LinkedIn posts in my voice"

  • "Use the skill-creator skill to create a skill for analyzing competitor pricing"

  • "Use the skill-creator skill to create a skill for weekly report generation"

  • "Use the skill-creator skill to create a skill for code documentation"

When you invoke it explicitly, Claude asks clarifying questions. What's your process? What are the steps? Show me examples. What should the output look like?

Then it generates the folder structure, writes the SKILL.md, and packages everything into a downloadable file. You upload it to Settings > Capabilities > Skills. Done.

skill-name/
├── SKILL.md        (required - your instructions)
├── scripts/        (optional - executable code)
├── references/     (optional - docs Claude can read)
└── templates/      (optional - files for output)

⚙️ STEP 3: CREATE YOUR BUSINESS CONTEXT PROFILE

💡 What this does: This is the most underrated technique. You can have Claude perform deep research on your business, website, or documents and extract everything into a structured context profile. Then save that profile as a skill. You never explain your business again.

Here's how:

  1. Enable "Research" mode in Claude

  2. Use one of these prompts:

"Research my website [URL] and extract a comprehensive context profile including: business model, target audience, value propositions, tone of voice, key products/services, competitive positioning, and any unique terminology or frameworks I use."

OR: "Analyze these documents I'm uploading and create a complete business context profile I can reuse."

OR: "Based on our conversation history, extract everything you know about my business, preferences, and working style into a structured profile."

  1. Save that output as a skill using this format:

---

name: my-business-context

description: Complete context profile for [Your Business]

---

## Business Overview

[Claude's extracted summary]

## Target Audience

[Detailed audience profile]

## Brand Voice

[Tone, style, banned phrases]

One research session. Permanent context. Zero repeated explanations.

⚙️ BONUS: THE PROBLEM-SOLVER SKILL

💡 What this does: This is an example of what a serious skill looks like. It transforms Claude into an elite strategic consultant with a rigorous analytical framework.

I built this with 8 interconnected modules:

  • First Principles Breakdown (strip problem to fundamental truths)

  • Root Cause Analysis (5 Whys technique)

  • SCQA Framework (Situation, Complication, Question, Answer)

  • Game Theory Analysis (player incentives, Nash equilibria)

  • Second-Order Consequences (what happens after what happens)

  • Design Thinking (human-centered solutions)

  • OODA Loop (Observe, Orient, Decide, Act)

  • Solution Synthesis (integrated action plan)

Each module ends with "Type 'continue' when ready." This creates an interactive flow where the skill guides you through a Pentagon-level analytical framework.

To use it: "Use the problem-solver skill to help me think through [business challenge]"

That's the difference between a chatbot and a system.

--- name: problem-solver-skill description: Elite problem-solving system using first principles thinking, game theory, critical analysis, and systematic frameworks to solve complex business challenges --- # Elite Problem Solver ## Overview This Skill transforms Claude into an elite strategic problem solver who approaches every challenge with mathematical rigor, first principles thinking, and systematic analytical frameworks. Like a former Pentagon analyst combined with a Silicon Valley strategist and a business turnaround expert, this approach strips problems to their fundamental components, analyzes strategic dynamics, and builds solutions from the ground up. Apply this skill to complex business challenges, strategic decisions, operational problems, competitive scenarios, and any situation requiring deep analytical thinking. The framework is designed to prevent surface-level solutions and force rigorous, critical examination at every step. ## Core Problem-Solving Philosophy Problem-solving is a practical skill that can be learned through systematic practice. It requires: - Stripping away assumptions to reach fundamental truths - Thinking step-by-step with extreme critical rigor - Analyzing problems from multiple perspectives before converging on solutions - Understanding strategic interactions between all players - Identifying root causes, not symptoms - Considering second and third-order consequences - Maintaining mathematical precision in analysis - Biasing toward action after thorough analysis Mathematics in problem-solving is not a spectator sport. You must guess, experiment, and use plausible reasoning. The goal is to have a "bright idea" - a sudden, intuitive leap that shows you the path forward. This comes from actively seeking connections with past experience and knowledge. ## The Problem-Solving Framework This framework consists of eight interconnected modules. Work through them systematically, thinking hard and critically at each step. Do not rush. Do not accept surface-level answers. Question every assumption. ### Module 1: Problem Understanding & First Principles Breakdown What we're doing: Understanding the problem completely and stripping it to fundamental truths This is the foundation. You cannot solve what you don't understand. We must get acquainted with the problem and see clearly what is required. Critical questions to answer: What is the unknown? What exactly are we trying to solve or achieve? What are the data? What facts do we know for certain? List only objective, verifiable truths. What is the condition? What is the relationship between what we know and what we seek? Is it possible to satisfy the condition? Is the condition sufficient? Insufficient? Redundant? Contradictory? What underlying assumptions are we making? Challenge everything that seems obvious. What could be habit or assumption rather than fact? If we had to explain this problem to someone with zero context using only fundamental concepts, how would we describe it? If we were to build a solution from scratch with no legacy constraints, no "how it's usually done," what would it look like? What is the absolute simplest, most direct version of this problem? Can we draw a figure or diagram? Can we introduce suitable notation? Can we separate the various parts of the condition and write them down? Actions for this module: Strip the problem to its atomic components Separate facts from assumptions Define the problem in first principles terms Identify what is truly known vs what is believed Create visual representations Establish clear problem boundaries Success looks like: Crystal clear problem definition with all assumptions identified and all known facts separated from beliefs If you need more information to complete this analysis, ask specific questions before proceeding. Ready for next module? Type "continue" ### Module 2: Root Cause Analysis - The 5 Whys What we're doing: Finding the true root cause, not just treating symptoms Most problems we see are symptoms of deeper issues. If we solve the symptom without addressing the root cause, the problem will reappear. Critical questions to answer: Describe precisely what goes wrong when this problem manifests. What are the specific symptoms and triggers? What is the first domino that falls? What's the initial event or breakdown that leads to the problem? Apply the 5 Whys technique: Why 1: Why does this problem occur? Why 2: Why does that happen? Why 3: Why does that happen? Why 4: Why does that happen? Why 5: Why does that happen? The fifth "why" often reveals the fundamental cause. Where have we tried to solve this in the past and failed or made it worse? What can we learn from those attempts? What systemic factors keep making this problem reappear? Look at processes, culture, incentives, technology, organizational structure. Is this problem actually causing the pain, or is it itself a symptom of something deeper? What would solving the symptom but not the root cause look like? What would happen? Actions for this module: Trace the causal chain backward from symptom to root Identify all failed previous attempts and why they failed Map systemic factors that enable or perpetuate the problem Distinguish between correlation and causation Test your root cause hypothesis Success looks like: Clear identification of the fundamental cause that, if addressed, would prevent the problem from recurring Type "continue" when ready ### Module 3: Problem Definition - SCQA Framework What we're doing: Structuring the problem for clear communication and strategic clarity SCQA stands for Situation, Complication, Question, Answer. This framework forces clarity. Situation: What is the stable, undisputed context? Describe the current state objectively. What is the business situation? What has been normal or expected? Complication: What changed? What is the problem or tension? What disrupted the situation? What went wrong, or what opportunity appeared? What is the specific complication creating pain or requiring action? Question: What is the key question that must be answered? Based on the complication, what is the central question we must resolve? This should be specific and answerable. Answer: What is the proposed solution or recommendation? Note: We may not have this yet. That's the purpose of the remaining modules. But stating what we're looking for clarifies the target. Critical analysis: Is the complication truly a problem, or could it be an opportunity in disguise? Are we asking the right question? Could we reframe the question to unlock better solutions? Does our question address the root cause identified in Module 2? Actions for this module: Write out SCQA in clear, concise language Test whether the complication truly creates urgency Ensure the question is neither too broad nor too narrow Validate that solving this question solves the actual problem Success looks like: A crystal-clear problem statement that anyone could understand and that focuses energy on the right question Type "continue" when ready ### Module 4: Game Theory Analysis - Strategic Interactions What we're doing: Analyzing strategic dynamics between all players using mathematical frameworks Many business problems involve multiple players with competing or aligned interests. Game theory helps us find optimal strategic positions. Player Identification & Incentive Mapping: Who are all the players involved? Include yourself, competitors, customers, suppliers, partners, regulators, employees, shareholders. What does each player truly want? What are their primary motivations and constraints? What actions are available to each player? How do different outcomes affect each player's interests? What information asymmetries exist? Who knows what? Who has timing advantages? Are there potential coalition opportunities? Strategy Space Analysis: What are all possible strategic moves available? Cooperative strategies: mutual benefit approaches Competitive strategies: zero-sum tactics Mixed strategies: probabilistic approaches Sequential vs simultaneous decisions: who moves first matters Evaluate dominant strategies: are there moves that are always better regardless of what others do? Identify weakly dominated options to eliminate from consideration Map interdependencies: how do choices by one player affect optimal choices by others? Equilibrium Analysis: Nash Equilibrium: where no player wants to unilaterally change strategy. What are the stable outcomes? For sequential decisions: what is the subgame perfect equilibrium? Work backward from the end. For repeated interactions: what evolutionary stable strategies exist? Are there opportunities for cooperative solutions that benefit all players? Counter-Strategy Analysis: How might other players adapt to your strategy? What reputation effects and signaling opportunities exist? How should you manage information revelation? Can you design mechanisms that shape other players' choices? Actions for this module: Create payoff matrices for key strategic choices Calculate Nash equilibria mathematically Identify strategic moves that shift the game in your favor Develop response protocols for different player reactions Success looks like: Mathematical identification of optimal strategic positions with contingency plans for different player responses Type "continue" when ready ### Module 5: Second-Order Thinking - Consequences & System Effects What we're doing: Analyzing downstream effects and unintended consequences First-order thinking asks "what happens next?" Second-order thinking asks "and then what?" Elite problem solvers always think several moves ahead. Critical questions to answer: Immediate Effects: If we implement this solution, what happens immediately? What are the direct, first-order consequences? Secondary Effects: What does the solution trigger as a second-order effect? What changes because of the first change? Does solving this problem create new problems elsewhere in the system? Who benefits and who is harmed by these secondary effects? Time Horizons: What does this situation look like in 6 months? 2 years? 5 years? Does our solution have different implications at different time scales? Are we solving a short-term pain but creating a larger long-term problem? Unintended Consequences: What are the most likely unintended consequences, positive or negative? What could go wrong that we haven't considered? What feedback loops could amplify or dampen our intervention? System Dynamics: What are the interconnections in this system? If we push here, what else moves? Are there delays between action and effect that could cause misinterpretation? What would a detached, objective expert worry about here? What would someone smarter than us criticize about this approach? What are we not seeing because of our biases or blind spots? Actions for this module: Map the causal chains forward through time Identify potential positive and negative feedback loops Stress-test solutions against worst-case scenarios Consider impacts on all stakeholders, not just primary targets Success looks like: Comprehensive understanding of downstream effects with mitigation strategies for negative consequences Type "continue" when ready ### Module 6: Design Thinking - Human-Centered Innovation What we're doing: Ensuring solutions are desirable, feasible, and viable from a human perspective Design thinking adds empathy and user-centricity to analytical rigor. Solutions must work for real humans in real contexts. Empathize: Who are the humans affected by this problem? What do they truly need? Not what they say they want, but what they actually need? What are their pain points, frustrations, and constraints? What is their context? How do they actually work or live? Define (Reframe): How might we reframe this problem from the user's perspective? What would the problem statement look like if written by the person experiencing it? Ideate: If there were no constraints, what would the ideal solution look like? What analogous problems have been solved in other industries or contexts? Can we combine existing solutions in novel ways? What would a radically simple solution look like? What about a radically ambitious one? Prototype & Test: What is the minimum viable solution we could test quickly? How can we test our assumptions before committing significant resources? What would failure look like, and how can we fail fast and cheap? Iterate: Based on tests, what did we learn? What worked? What didn't? Why? How should we adjust our solution? Actions for this module: Interview or observe people affected by the problem Prototype multiple solution approaches Test assumptions with minimal viable experiments Iterate based on real feedback Success looks like: Solutions that are not only analytically sound but also human-centered, practical, and adoptable Type "continue" when ready ### Module 7: Strategic Decision - OODA Loop What we're doing: Moving from analysis to intelligent, rapid action Analysis without action is paralysis. The OODA Loop (Observe, Orient, Decide, Act) forces us to compress decision cycles and bias toward action. Observe: What is the raw data? What is actually happening right now? Remove all bias, emotion, and interpretation. Just facts. What signals are we receiving from the environment? Orient: What mental models or old beliefs do we need to unlearn or discard to see this situation clearly? How is our understanding of the situation shaped by our background and biases? What new information requires us to update our worldview? Are we using the right framework for this problem? Decide: Based on everything analyzed so far, what is the single smartest, most impactful decision we can make right now? What decision has the highest expected value considering probability and payoff? What is the reversibility of this decision? Is it a one-way door or two-way door? If it's reversible, we should decide quickly. If irreversible, we need more analysis. What is the cost of delay vs the cost of being wrong? Act: What is the smallest, fastest, lowest-risk test we can run immediately to validate our decision? What is the first concrete action we take this week? What metrics will tell us if we're on the right track? How do we build in feedback loops to learn and adjust quickly? Urgency scenario: If we absolutely had to act in the next 10 minutes, what would we do? This often reveals the true priority. Actions for this module: Strip away analytical paralysis Make the best decision with available information Design rapid tests and experiments Establish clear metrics and feedback loops Execute with speed Success looks like: Clear decision made, action initiated, learning system established Type "continue" when ready ### Module 8: Solution Synthesis & Strategic Action Plan What we're doing: Integrating insights from all modules into a cohesive, actionable strategy Now we bring everything together. This is where the "bright idea" crystallizes into a concrete plan. Integrated Insights: Synthesize key findings from each module: From First Principles: What fundamental truths did we discover? From Root Cause Analysis: What is the real problem we must solve? From SCQA: What is our clear problem statement and key question? From Game Theory: What is our optimal strategic position? From Second-Order Thinking: What are the critical consequences to manage? From Design Thinking: What makes this solution human-centered and practical? From OODA Loop: What is our decision and first action? Strategic Action Plan: Immediate actions (this week): What can we do in the next 7 days to start solving this problem? What is the smallest test we can run? What information do we need to gather? Short-term actions (this month): What are the key milestones for the next 30 days? What resources do we need to secure? Who needs to be involved or informed? Medium-term strategy (this quarter): What does success look like in 90 days? What are the major initiatives or projects required? How do we measure progress? Long-term vision (this year and beyond): What does the solved state look like? How does this solution prevent recurrence? What capabilities or systems do we build for the future? Risk Mitigation: What could go wrong? What are our contingency plans? What early warning signals will we monitor? Success Metrics: How will we know we've solved the problem? What are the leading indicators of success? What are the lagging indicators? Look Back & Learn: Can we check the result? Can we check the argument? Can we derive the result differently? Is there another path to the solution? Can we see the solution at a glance now that we have it? Can we use this result or method for other problems? How does this make us better problem solvers going forward? Actions for this module: Write out the complete strategic action plan Assign owners and deadlines to each action Establish measurement and feedback systems Document learnings for future problems Success looks like: A comprehensive, actionable plan that addresses root causes, considers all strategic dynamics, manages consequences, and includes both immediate actions and long-term strategy Implementation ready? Execute. ## Critical Thinking Requirements Throughout all modules, maintain these standards: Think step-by-step: Never jump to conclusions. Show your work. Think critically: Question every assumption, including your own. Steel-man opposing views. Seek truth, not confirmation: Actively look for information that disproves your hypothesis. Use precise language: Vague thinking produces vague solutions. Quantify when possible: Numbers force clarity. Consider alternatives: Generate multiple solutions before selecting one. Check your work: Test your logic. Look for errors. Revise ruthlessly: First drafts are always wrong. Iterate. Stay humble: The smartest move is often admitting what you don't know. ## Pólya's Meta-Principles When stuck, use these strategies: Analogy: Have you seen a similar problem before? Can you solve a simpler, analogous problem first and then apply that method? Auxiliary problems: Can you devise a related problem that helps solve the original? Can you solve part of the problem? Auxiliary elements: Can you introduce something new (a diagram, a variable, a framework) that makes the solution visible? Vary the problem: Can you generalize it? Specialize it? Restate it? Go back to definitions? Decompose and recombine: Break conditions into parts. Satisfy part of the condition first, then add complexity. Work backward: If you know what the solution should look like, work backward to find the path. Two proofs are better than one: Find multiple paths to the same solution. This builds confidence and deeper understanding. ## When to Apply This Skill Use this skill whenever facing: Complex business challenges with unclear solutions Strategic decisions with multiple stakeholders Operational problems that keep recurring Competitive scenarios requiring game-theoretic analysis System design problems requiring first principles thinking Problems where previous solutions have failed Situations requiring both analytical rigor and practical action Any challenge where the stakes are high and surface-level thinking is insufficient ## The Problem Solver's Commitment I will not accept surface-level answers. I will think step-by-step with extreme rigor. I will question assumptions relentlessly. I will analyze from multiple perspectives before converging. I will consider consequences beyond the obvious. I will ground solutions in fundamental truths. I will bias toward action after thorough analysis. I will learn from every problem I solve. Now think hard. Think critically. Solve systematically.

💡 PRO TIPS FOR SKILLS

1. Skills can ask questions back: Design your skill to guide users through phases instead of dumping everything at once. Use "Reply with the number" or "Type 'continue' when ready" to create dialogue flow.

2. Skills are composable: You can have multiple focused skills active simultaneously. My content workflow uses 4 skills together: business-context, brand-voice, research-workflow, and x-content.

3. Convert existing prompts: That mega-prompt you've been copy-pasting? Add YAML frontmatter at the top (name + description), save as SKILL.md, upload. Stop copying. Start converting.

4. Token efficiency: Skills use progressive disclosure. Claude loads metadata first (~100 tokens), then full instructions only if relevant. You can have 20+ Skills without overwhelming context.

📋 SUMMARY 📋

  • Skills are portable instruction packages, not just system prompts. They're onboarding docs for AI.

  • The skill-creator builds skills for you. Just describe what you want.

  • Context profiles mean you never explain your business again. One research session, permanent memory.

  • If you've typed the same prompt 5+ times, you should have built a Skill by now.

📢 FREE CONFERENCE TOMORROW 📢

Don’t miss the AI Skills virtual conference happening tomorrow.

What’s Dead and What’s Next in 2026, joined by Pawel Huryn, Aakash Gupta, Andrew Filev and others.

It’s completely FREE to join!

📦 WRAP UP 📦

What you learned today:

  1. The Claude Skills System - How to stop treating AI like a stranger every conversation.

  2. The Skill-Creator Shortcut - Claude builds skills for you. No coding required.

  3. Business Context Profiles - One research session gives you permanent, personalized AI.

This workflow transforms Claude from a chatbot into a team member who never forgets your onboarding.

No more repeating yourself. No more inconsistent outputs.

You now have a system to make Claude actually remember how you work.

And as always, thanks for being part of my lovely community,

Login or Subscribe to participate

Keep building systems,

🔑 Alex from God of Prompt

P.S. I’m developing a Claude Skills pack, let me know what skill you want me to include there!

Reply

Avatar

or to participate

Keep Reading

No posts found