freekmurze/dotfiles

The ai-seo skill is designed for optimizing content to be discoverable, extractable, and citable by AI systems like Google AI Overviews, ChatGPT, Perplexity, Claude, Gemini, and Copilot.

When to use this skill:

• When you want to optimize for 'AI SEO', 'AEO' (Answer Engine Optimization), 'GEO' (Generative Engine Optimization), or 'LLMO' (LLM Optimization).
• When you want to appear in AI-generated answers or citations.
• When you want to optimize for 'AI Overviews', 'ChatGPT', or 'Perplexity'.

Related skills:

• For traditional technical/on-page SEO audits, use seo-audit.
• For structured data implementation, use schema-markup.

Use the popup-cro skill when you need to create or optimize popups, modals, overlays, slide-ins, or banners for conversion purposes. This includes optimizing for 'exit intent,' 'lead capture,' 'email popups,' or 'announcement banners.'

Note: For optimizing the forms inside a popup, use form-cro. For general page conversion optimization, use page-cro.

The churn-prevention skill is designed to help reduce both voluntary churn (customers choosing to cancel) and involuntary churn (failed payments).

Core Strategies

• Voluntary Churn: Managed through cancel flows, dynamic save offers, and exit surveys.
• Involuntary Churn: Managed through dunning emails, smart retries, and card updaters.

Supported Modes

1. Build a cancel flow: Designing a sequence from scratch including surveys and offers.
2. Optimize an existing flow: Analyzing data to improve save rates.
3. Set up dunning: Implementing failed payment recovery strategies.

For post-cancel win-back sequences, refer to the email-sequence skill. For in-app upgrade paywalls, refer to paywall-upgrade-cro.

The review-pr skill is designed to review and merge GitHub pull requests for Spatie packages. It automates the workflow of fetching PR details, reviewing diffs, checking CI status, and performing merges and releases using the gh CLI.

Trigger this skill by asking to:

• "review a PR"
• "review a pull request"
• "merge a PR"
• "review this PR"
• "check this pull request"
• "merge this"
• Use the command /review-pr
• Provide a GitHub PR URL.

The skill operates through a high-fidelity engineering pipeline consisting of five stages:

1. Retrieval: Uses a system-level curl script to bypass TLS/SNI issues on Google Cloud Storage.
2. Mapping: Cross-references Stitch metadata with the local style-guide.json to ensure token consistency.
3. Generation: Scaffolds components using a strict Atomic Design pattern.
4. Validation: Runs an automated AST check using @swc/core to prevent hardcoded hex values or missing interfaces.
5. Audit: Performs a final self-correction check against a 20-point architecture checklist.

When designing paywall triggers, experiment with the timing of the prompt and the mechanism used to restrict access.

Timing Strategies

• Engagement-based: Trigger after an 'aha moment' or after a specific number of actions/days.
• Reminder-based: Use early trial reminders (e.g., 7 days) versus late reminders (e.g., 1 day before expiration).
• Usage-based: Trigger based on specific usage patterns rather than just time elapsed.

Trigger Mechanisms

• Hard Gate: Prevent the user from proceeding until they upgrade.
• Soft Gate: Allow a preview of the feature followed by a prompt.
• UI Patterns: Test in-context modals, dedicated upgrade pages, banners, or exit-intent prompts on free plan pages.

Name exceptions after the specific failure and provide static factory methods. This allows for clean, readable error handling and provides helpful context to the user.

class CouldNotGeneratePdf extends Exception
{
    public static function browsershotNotInstalled(): static
    {
        return new static('To use Browsershot, install it via `composer require spatie/browsershot`.');
    }
}

Optimize how customers perceive and respond to costs using these models:

• Charm Pricing / Left-Digit Effect: Use prices ending in .99 or .95 for value-focused products to exploit the dominance of the left digit.
• Rounded-Price (Fluency) Effect: Use round numbers (e.g., $100) for premium products to signal quality; use charm prices for value products.
• Rule of 100:
  • For items under $100: Use percentage discounts (e.g., "20% off").
  • For items over $100: Use absolute discounts (e.g., "$50 off").
• Price Relativity / Good-Better-Best: Present three tiers where the middle tier is the target; the expensive tier acts as an anchor and the cheap tier provides a baseline.
• Mental Accounting: Reframe prices into smaller, more digestible units (e.g., "$1/day" instead of "$30/month" or "Less than your morning coffee").

For advanced micro-frontend architectures, Re.Pack supports Module Federation. This allows a 'Host' application to load modules from a 'Remote' application at runtime. This enables independent deployments and shared dependencies across different teams.

// Host app loads remote module
const RemoteModule = React.lazy(() =>
  import('remote-app/Module')
);

Use square brackets to create dynamic segments in your file paths.

Dynamic Routes Matches specific segments like /users/123:

• app/users/[id].tsx
• app/users/[id]/posts.tsx (matches /users/123/posts)

Catch-All Routes Use [...slug] to match any number of path segments:

• app/docs/[...slug].tsx (matches /docs/a, /docs/a/b, etc.)

app/
  users/
    [id].tsx        # Matches /users/123
    [id]/
      posts.tsx     # Matches /users/123/posts

  docs/
    [...slug].tsx   # Matches /docs/a/b/c

To avoid overspending, calculate your maximum allowable reward using the following formula. This ensures the incentive stays within your Customer Lifetime Value (LTV) and target Customer Acquisition Cost (CAC) constraints.

Formula: Max Referral Reward = (Customer LTV × Gross Margin) - Target CAC

Example Calculation:

• LTV: $1,200
• Gross margin: 70% (0.70)
• Target CAC: $200
• Max reward: ($1,200 × 0.70) - $200 = $640