Good morning.
Too many entrepreneurs still use ChatGPT or Claude like a search engine with better answers.
Ask a question, get a response, start over.
Skills change the paradigm. It’s a new form of prompting where you’re, basically, deploying expertise to get stuff done.
A Skill is a markdown file that gives Claude specialized knowledge for specific tasks. When you install marketing skills, Claude recognizes when you're working on positioning, competitor analysis, CRO, SEO, or pricing and automatically applies the right frameworks.
The methodology is baked in. The output structure is consistent and the expertise compounds over time.
Over the next few issues, we're building a complete marketing skill library across several domains:
Strategy, CRO, SEO, paid ads, measurement, ideation, and pricing.
This issue is your operating manual. You'll learn how skills work, how to install them, how to customize them, and how to run implementation projects that produce real results. Everything else in this series builds on what you learn here.
Yes, this requires Claude. Frankly, Claude is superior to ChatGPT at this point. I happily pay for the Max plan and suggest you do, too.
— Sam
IN TODAY’S ISSUE 🤖
What Claude Skills actually are (and what they're not)
Why this matters for entrepreneurs building marketing systems
How to install and organize your Skills
The mechanics: auto-triggering, manual invocation, and Skill chaining
The implementation framework you'll use for every issue
The Strategy & Psychology stack: five foundational Skills
Your first project: the positioning audit
Download the Skills
Let’s get into it.
1. What Claude Skills Are (And Aren't)
A Claude Skill is a markdown file. That's it.
The file contains three things:
YAML frontmatter at the top (name, description, trigger conditions)
Instructions Claude follows when the Skill activates
Frameworks, checklists, examples, or output formats that guide the work
When you install Skills, Claude scans the frontmatter to determine relevance. This uses roughly 100 tokens per Skill. When a Skill triggers, the full instructions load into context (typically under 5,000 tokens). Resources and examples load only as needed.
Multiple Skills can be available without overwhelming Claude's context window. Claude decides which Skill applies based on your request and the Skill descriptions.
As I mentioned earlier, this is for now used inside Claude. Perhaps at some point, ChatGPT will also have a Skills feature.
You can get a similar result from ChatGPT if you add the Skills content to the custom instructions of a Project, but it won’t behave the same. Still, it sort of works. Which means you could actually use some of this if you’re, for some reason, only using ChatGPT and refuse to use Claude.
My recommendation is that you get at least the Pro plan (or better, the Max plan, which is a huge unlock right now) of Claude. At this point, it’s beating ChatGPT across the board. I don’t see a good reason to keep ChatGPT for a business (at this time, that might change).
Skills are like portable expertise. Install once, available everywhere.
How Skills Differ from Other Claude Features
Feature | What It Does | Limitation |
Custom Instructions | Persistent personality/preferences | Same for every conversation |
Projects | Organized workspaces with knowledge | Manual setup per project |
Prompts | One-time instructions | Disappear after use |
Skills | Specialized expertise that auto-triggers | Requires installation |
Skills fill the gap between these. They're specialized expertise that auto-triggers based on context. They're portable across conversations. They encode methodology, not just information. And they compound: refine a Skill once, every future use benefits.
What Skills Are Good For
Repeatable frameworks you use across multiple contexts
Domain expertise that requires specific methodology
Workflows with multiple steps that benefit from consistency
Output formats that need to follow particular structures
What Skills Aren't Good For
One-off tasks with unique requirements
Creative work that should vary each time
Simple questions that don't need frameworks
Tasks where you want Claude to approach fresh
The rule of thumb: if you've explained the same process to Claude more than five times, it should be a Skill.
2. Why This Matters for Entrepreneurs
Growing businesses have marketing departments. Specialists who own positioning, CRO, SEO, paid media, and analytics.
Most entrepreneurs don't, or maybe they have a small team or one person trying to fill three or more roles. Whether you're running lean by choice or building toward scale, you're often making marketing decisions without dedicated specialists in the room. This is true even if you have a team.
Skills close the gap. Each Skill is a specialist who:
Knows their domain deeply
Follows proven methodology
Produces consistent outputs
Never needs retraining
Works for you and your team
If you had, for example, 34 Skills = 34 specialists available on demand.
I’m not saying you should have 34 Skills, by the way. That would be overkill.
The Compound Effect
Without Skills, every Claude session starts from zero. You re-explain your frameworks. You re-describe your methodology. You re-teach Claude how you think.
With Skills, knowledge accumulates. Refine a Skill once, every future use benefits. Your positioning framework improves, every positioning analysis improves.
This applies whether you're the one using Claude or you've got team members running these analyses. The methodology stays consistent.
Skills are institutional memory that scales with your team. They raise the floor on quality.
Your worst marketing analysis (tired, rushed, distracted) now follows the same framework as your best. Your newest team member's first CRO audit follows the same methodology yours does.
Consistency creates reliability. Reliability builds trust in the outputs and the process.
3. Installation & Setup
All Claude Skills are installed in one place: Settings > Capabilities.
Go to Settings, find Capabilities and scroll down for Skills
Installing Skills
Go to Settings (click your profile icon)
Select Capabilities
Under Skills, click Add Skill
Upload the
SKILL.mdfileRepeat for each Skill
Once installed, Skills are available across your account. They work in regular conversations and in any Claude Projects you create.
Verifying Installation
After installation:
Start a new conversation
Type: "What marketing Skills do you have available?" or
/name-of-skillto call upon a Skill you know is installed (obviously, you replace name-of-skill with the actual Skill name)Claude should list installed Skills with descriptions
If Skills don't appear:
Check the file uploaded correctly
Verify YAML frontmatter is valid
Try refreshing the page
Using Skills with Claude Projects
Skills provide methodology. Projects provide context.
After installing your Skills, create Claude Projects to organize your work by domain. Projects give you:
Dedicated workspace for specific work streams
Project-specific knowledge (upload relevant docs, data, context)
Conversation history within that domain
Team sharing (on Team/Enterprise plans)
Recommended Project Setup
Project Name | Skills You'll Use | Knowledge to Upload |
Strategy & Positioning | marketing-psychology, offer-positioning, competitor-analysis, customer-research, market-sizing | Brand guidelines, competitor research, customer testimonials |
Website Optimization | CRO stack (upcoming issue) | Page URLs, analytics data, heatmaps |
Content & Search | SEO/GEO stack (upcoming issue) | Content inventory, keyword research, site structure |
Advertising | Paid Ads stack (upcoming issue) | Ad accounts, landing pages, audience data |
Analytics | Measurement stack (upcoming issue) | GA4 access, conversion data, test history |
Revenue & Pricing | Pricing stack (upcoming issue) | Current pricing, competitor pricing, customer segments |
The Skills live in Settings. The context lives in Projects. Together, they create specialized workspaces where Claude has both the methodology and the information to do meaningful work.
4. How Skills Work
Understanding Skill mechanics helps you get more from them.
Anatomy of a Skill File
Every Skill follows the same structure. At the top, YAML frontmatter defines the Skill's name, description, and when it should trigger. The description is critical because it tells Claude when to activate the Skill.
Below the frontmatter, markdown content provides instructions Claude follows when the Skill is active. This typically includes the methodology or framework, step-by-step process, output format specifications, and examples of good and bad outputs.
Here's what a simplified Skill looks like:
---
name: offer-positioning
description: Framework for articulating what you sell, to whom, and why you win. Use when working on positioning, value propositions, or competitive differentiation.
---
# Offer Positioning
[Instructions Claude follows when this Skill activates]
## Framework
[The methodology]
## Process
[Step-by-step approach]
## Output Format
[What the deliverable looks like]
## Examples
[Good and bad examples]The description field does the heavy lifting. Be specific about use cases so Claude knows when to activate the Skill.
Auto-Trigger vs. Manual Invocation
Skills can activate two ways.
Auto-Trigger: Claude reads your request, scans Skill descriptions, activates relevant Skills automatically.
You ask: "Help me analyze my positioning against these three competitors" Claude thinks: "This involves positioning and competitor analysis" Skills activate: offer-positioning, competitor-analysis
Manual Invocation: You explicitly request a Skill.
You say: "Use the offer-positioning Skill to analyze my current value prop" Claude activates: offer-positioning regardless of whether it would have auto-triggered
When to use manual invocation:
Testing a new Skill
Forcing a specific methodology
Combining Skills that wouldn't naturally trigger together
Overriding Claude's Skill selection
Skill Chaining
Skills can trigger other Skills. A page-cro audit might identify form issues, triggering form-cro for deeper analysis.
Chaining happens when:
One Skill's output naturally leads to another Skill's domain
You explicitly ask Claude to "continue with [Skill name]"
The Skill instructions reference other Skills
Example chain: customer-research → offer-positioning → competitor-analysis
This is where Skills get powerful. You're not running isolated analyses. You're running connected workflows where insights from one Skill feed into the next.
5. Customizing Skills
The Skills in this series work out of the box. But you can customize them for your specific context.
When to Customize
Customize when:
Your industry has specific terminology
Your business model requires different frameworks
You've developed proprietary methodology
Default outputs don't match your needs
Don't customize when:
You haven't used the default version enough to know what's missing
The change is one-time (just include it in your prompt)
You're optimizing prematurely
How to Customize
Download an example
SKILL.mdfile (or just create a blank file)Make a copy (preserve the original)
Edit the copy
Upload the customized version
Test with real tasks
Iterate based on outputs
Customization Examples
You might add industry context to a Skill:
## Industry Context
This framework applies to B2B SaaS companies selling to enterprise buyers. Adjust language and examples accordingly. Common objections in this space include: security concerns, integration complexity, and switching costs.You might modify the output format:
## Output Format
Deliver as a one-page brief with:
- Executive summary (3 sentences max)
- Key findings (bullet points)
- Recommendations (prioritized list)
- Next steps (specific actions with owners)You might add your own frameworks:
## Positioning Framework
Use the CORTEX positioning canvas:
1. Customer: Who specifically (title, company size, situation)
2. Problem: What pain, quantified
3. Solution: What you do, simply stated
4. Differentiator: Why you vs. alternatives
5. Proof: Evidence it worksName customized versions clearly. Something like offer-positioning-saas.md makes it obvious this is your SaaS-specific version. When Skill updates ship, you can compare the new version to your customizations and merge the changes you want.
When to Use Default vs. Custom
Start with defaults. Customize only after five or more uses of the default version, when you have a clear pattern of what's missing, and when you have a specific improvement in mind.
Premature customization creates maintenance burden without proven benefit. Let the defaults prove themselves insufficient before you invest time in modifications.
6. The Implementation Framework
This is the approach you'll use for every issue in this series. Each Skill stack comes with a suggested project that takes four to six hours.
The Four-Phase Approach
Phase 1: Audit (1.5-2 hours)
Run relevant Skills against your current state. Don't fix anything yet. Just document.
Questions to answer:
What does the Skill reveal about current performance?
Where are the gaps between current and optimal?
What patterns emerge across multiple Skill outputs?
Phase 2: Prioritize (30 minutes)
You'll have more findings than you can address. Prioritize using:
Impact: How much will fixing this move the needle?
Effort: How long will implementation take?
Confidence: How sure are you this will work?
Top 3-5 items become your focus.
Phase 3: Implement (2-3 hours)
Execute on prioritized items. Use Skills to guide implementation, not just analysis.
For each item:
Define the specific change
Make the change
Document what you did
Phase 4: Measure (Ongoing)
Before you start, define what you'll track. After implementation, monitor.
Minimum viable measurement:
What metric indicates success?
What's the current baseline?
When will you check progress?
What Done Looks Like
A completed project produces:
Audit document with findings
Prioritized list of opportunities
3-5 implemented changes
Measurement plan with baselines
You're not trying to fix everything. You're trying to make meaningful progress with clear documentation.
Documenting for Future Reference
Create a simple log for each project:
Date
Skills used
Key findings
Changes made
Results (updated over time)
This becomes your institutional memory. Six months from now, you'll know what you tried, what worked, and what didn't.
7. Testing & Refinement
Skills improve through use. Here's how to know if they're working and how to make them better.
How to Know if a Skill Is Working
A Skill is working when:
It activates on relevant requests (auto-trigger functions)
Outputs follow the intended framework
Results are useful without heavy editing
You use it repeatedly
A Skill needs refinement when:
It triggers when it shouldn't (or doesn't trigger when it should)
Outputs miss key elements
You consistently edit the same things
You avoid using it
The Refinement Loop
Use the Skill on a real task
Evaluate the output against your needs
Identify gaps between output and ideal
Edit the Skill to address gaps
Test again with a similar task
Repeat until outputs consistently meet standards
Most Skills need two or three rounds of refinement before they're dialed in. That's normal. The investment pays off across every future use.
Common Fixes
Problem | Fix |
Skill doesn't trigger | Expand description with more use cases |
Skill triggers incorrectly | Make description more specific |
Output misses key sections | Add explicit output format requirements |
Wrong level of detail | Specify depth in instructions |
Generic language | Add examples of good outputs |
Inconsistent frameworks | Make methodology steps explicit |
8. The Strategy & Psychology Stack
This issue delivers five foundational Skills. These inform everything tactical that follows. Strategy before tactics. Always.
Skill 1: marketing-psychology
What it does: Applies mental models, cognitive biases, and persuasion frameworks to your marketing challenges.
When it triggers: Questions about why customers behave certain ways, how to increase persuasion, what psychological principles apply.
Key frameworks included:
Loss aversion and framing
Social proof mechanics
Anchoring and adjustment
Reciprocity loops
Commitment and consistency
Scarcity and urgency (ethical applications)
Example use: "Why aren't visitors converting on my pricing page? What psychological factors might be at play?"
Here’s an example of what the SKILL.md file would look like:
---
name: marketing-psychology
description: Apply behavioral psychology and persuasion principles to marketing challenges. Use for conversion optimization, pricing psychology, copy improvement, and understanding customer behavior.
---
# Marketing Psychology
Apply behavioral psychology frameworks to diagnose and improve marketing effectiveness. This skill helps you understand why customers behave the way they do and how to ethically influence purchase decisions.
## When to Use This Skill
- Diagnosing why a page, email, or offer isn't converting
- Designing pricing presentations and tier structures
- Writing copy that addresses emotional drivers
- Building trust and reducing purchase anxiety
- Understanding objections before they surface
- Analyzing competitor psychological tactics
## Core Psychological Principles
### 1. Loss Aversion
People feel losses roughly twice as intensely as equivalent gains.
**Application:**
- Frame benefits as what customers lose by not acting
- Highlight costs of the status quo
- Use trial periods that create ownership before purchase
**Example:** "Stop losing $2,400/year to inefficient processes" vs. "Save $2,400/year"
### 2. Social Proof
People look to others' behavior to guide their own decisions, especially under uncertainty.
**Types (ranked by strength):**
1. Specific testimonials with names, photos, measurable results
2. Case studies with detailed outcomes
3. Aggregate numbers ("10,000+ customers")
4. Logos of recognizable clients
5. Star ratings and review counts
**Application:**
- Match social proof to prospect's identity (similar industry, role, company size)
- Use specifics over generics ("Increased revenue 43%" vs. "Great results")
- Place social proof near high-friction points (pricing, signup forms)
### 3. Anchoring
The first number people see influences all subsequent judgments.
**Application:**
- Show highest price tier first
- Display original price before discount
- Lead with largest benefit number
- Compare to more expensive alternatives
**Example:** "$10,000 value for just $997" anchors perception to $10,000.
### 4. Reciprocity
People feel obligated to give back when they receive something of value.
**Application:**
- Provide genuine value before asking for anything
- Free tools, templates, audits
- Personalized recommendations
- Educational content without gates
**Warning:** Perceived manipulation destroys reciprocity. Value must be genuine.
### 5. Commitment and Consistency
People want to act consistently with prior commitments, especially public ones.
**Application:**
- Get small yeses before big asks (micro-commitments)
- Use quizzes and assessments that create investment
- Ask for public commitments (sharing, testimonials)
- Progressive profiling over time
**Example:** Free account → Profile completion → First action → Upgrade prompt
### 6. Scarcity and Urgency
Limited availability increases perceived value.
**Ethical application:**
- Real deadlines (cohort starts, event dates)
- Genuine capacity limits (consulting spots, inventory)
- Time-limited bonuses (not the core offer)
**Unethical (avoid):**
- Fake countdown timers that reset
- "Only 3 left" that never changes
- False urgency without real constraints
### 7. Authority
People defer to perceived experts and credible sources.
**Application:**
- Display credentials relevant to the claim
- Cite research and data sources
- Show media mentions and awards
- Partner with recognized names
### 8. Unity
People favor those they perceive as part of their group.
**Application:**
- Use "we" language that creates shared identity
- Reference shared experiences, challenges, values
- Build community and belonging
- Speak the audience's language
## Diagnostic Framework
When conversion is underperforming, analyze through this lens:
### Step 1: Identify the Friction Point
Where are people dropping off?
- Page bounce
- Form abandonment
- Checkout exit
- Trial non-conversion
### Step 2: Diagnose the Psychological Barrier
| Symptom | Likely Barrier | Relevant Principle |
|---------|---------------|-------------------|
| High bounce, low scroll | Unclear value | Anchoring, Framing |
| Scroll but no click | Insufficient motivation | Loss Aversion, Scarcity |
| Click but no form complete | Trust concerns | Social Proof, Authority |
| Form start but abandon | Too much friction | Commitment (too big ask) |
| Trial but no convert | Value not demonstrated | Reciprocity, Consistency |
### Step 3: Apply Appropriate Principle
Match the principle to the barrier. Don't apply principles randomly—diagnose first.
## Output Format
When analyzing a marketing asset:
```
## Psychological Analysis: [Asset Name]
### Current State
- Primary conversion goal:
- Current conversion rate (if known):
- Drop-off point:
### Psychological Barriers Identified
1. [Barrier]: [Evidence from asset]
2. [Barrier]: [Evidence from asset]
### Recommended Changes
1. [Change]: Apply [Principle] by [specific action]
- Expected impact: [High/Medium/Low]
- Implementation complexity: [High/Medium/Low]
### Ethical Considerations
- [Any concerns about manipulation vs. persuasion]
```
## Examples
### Good Application
**Situation:** SaaS pricing page with 2% conversion
**Diagnosis:** No social proof near pricing, no anchoring, benefits framed as gains not losses
**Recommendations:**
1. Add customer logos and testimonial near pricing table (Social Proof)
2. Show "Most Popular" on middle tier with 60% of customers choosing it (Social Proof + Anchoring)
3. Reframe "Save 10 hours/week" as "Stop losing 520 hours/year" (Loss Aversion)
### Bad Application (Avoid)
- Fake scarcity ("Only 2 spots left!" that never changes)
- Manipulative urgency without real deadlines
- Cherry-picked or fabricated testimonials
- Dark patterns that trick rather than persuade
## Ethical Guidelines
Persuasion helps people make decisions aligned with their interests.
Manipulation tricks people into decisions against their interests.
**Test:** Would the customer thank you for this tactic if they knew about it?
- Yes → Persuasion (proceed)
- No → Manipulation (don't do it)
Always:
- Be truthful in all claims
- Deliver on promises made
- Make it easy to reverse decisions
- Respect customer autonomySkill 2: offer-positioning
What it does: Helps articulate what you sell, to whom, and why you win against alternatives.
When it triggers: Questions about positioning, value propositions, differentiation, messaging.
Key frameworks included:
Positioning canvas (customer, problem, solution, differentiator, proof)
Competitive positioning map
Value proposition hierarchy
Message-market fit assessment
Example use: "Help me refine my positioning for enterprise buyers. Here's my current value prop and my three main competitors."
Skill 3: competitor-analysis
What it does: Systematic framework for mapping competitive landscape and finding gaps.
When it triggers: Questions about competitors, market positioning, differentiation opportunities.
Key frameworks included:
Competitor identification (direct, indirect, substitutes)
Feature/benefit comparison matrix
Positioning gap analysis
Weakness exploitation framework
Example use: "Analyze how I stack up against these three competitors. Here are their websites and my current positioning."
Skill 4: customer-research
What it does: Mine voice-of-customer data from reviews, testimonials, support tickets, and interviews.
When it triggers: Questions about customer language, pain points, desires, objections.
Key frameworks included:
VOC extraction methodology
Pain/gain categorization
Jobs-to-be-done analysis
Review mining prompts
Example use: "What are customers really saying about solutions in my space? Here are 50 reviews from competitor products."
Skill 5: market-sizing
What it does: Estimate market opportunity using TAM/SAM/SOM methodology.
When it triggers: Questions about market size, opportunity assessment, new market entry.
Key frameworks included:
Top-down vs. bottom-up sizing
TAM/SAM/SOM definitions and calculations
Assumption documentation
Sensitivity analysis
Example use: "How big is the market for this new product idea? Here's what I know about the space."
Do you want ALL the Skills, ready to deploy?
You got the first one above, marketing-psychology. If you’re a Cortex subscriber, you’ll get them all below.
If you’re not a Cortex subscriber, you get full access when you join.
What is Cortex?
Cortex is the premium monthly membership version of Bionic Business.
In addition to 3-4 full, unrestricted regular issues per month, you also get:
1x SIGNALS Strategic Briefing issue (strategic moves you must make now to secure your business revenue, market share, and profits).
1x CIRCUITS Tactical Guide issue (workflows, how-to’s, prompts, very practical implementation inside a self-driving business).
50% OFF workshops on prompt engineering, agent operating systems, and more. Only active Cortex subscribers get any kind of discount, no one else.
For free subscribers, get a redacted and summarized version of regular issues (only Cortex subscribers get the full emails loaded with the good stuff). There’s enough value for you to take action but it’s not the full enchilada. And some issues are in full, with no restrictions.
So, if you’d like to receive full issues and the monthly special Signals and Circuits issues, then you should sign up for Cortex, which is $50/month (and you have full archives of previous regular issues for as long as you’re a member).
One note on Signals and Circuits: You only get these special issues for the months you’re subscribed—which means if you skip a month, you won’t get them for that month. There are no back-issues for past months, so now is your chance to lock-in these upcoming special issues while you can.
Also, almost all issues on the website are only accessible with a Cortex subscription.
After you’ve upgraded to Cortex, log in to the website and come back to this issue, and the Skills will be unlocked below.
Ready? Get the Skills, full regular issues, and the upcoming Signals and Circuits issues:
Cortex is a monthly premium version of Bionic Business, helping entrepreneurs grow and scale with AI and Agents. You can cancel at any time. But there are no refunds, for any reason. All sales are final.
Join Cortex to read the rest!
Become a paying subscriber to get access to this post and other subscriber-only issues. Cortex is a monthly premium version of Bionic Business, helping entrepreneurs and business owners grow and scale with AI and Agents—and build autonomous businesses.
Upgrade to Cortex NowEvery month, you get:
- Full regular email issues
- 1x Signals Strategic Briefing
- 1x Circuits Tactical Guide
