DTC Audit Skills Pack / Productized / AI Workflows
16 production skills. The next 4 ship free.
Twelve Claude Code skills that audit a DTC Shopify stack across tracking, retention, ops, and brand. Buy today, the next four skills land free as they ship in production.
// the problem
Claude Code is powerful but every operator writes their own skills from scratch. That is wasted hours. You want the working versions, not the getting-started versions. And you want them in a format you can version, share with your team, and evolve over time.
// the promise
A production pack of Claude Code skills covering the full DTC operator surface. Install once, use everywhere, get the next round of skills free as they ship.
// this is for you if
- You use Claude Code and are tired of re-writing the same prompts
- You run DTC marketing or Shopify work and want repeatable diagnostic patterns
- You are building an agent library for your team
- You want to learn skill authoring by studying production examples
- You buy the diagnostic products on this site and want to run the same modules in your own repo
// what’s inside
01 / Tracking + Attribution
Catch CAPI leaks before they cost you money. Reconcile what Shopify reports against what Meta and GA4 are crediting themselves.
- $capi-leak-scan-lite
- $attribution-reconciler
02 / Acquisition + Creative
Brief the next round of Meta creative against your real winners. Reallocate budget based on contribution margin, not last-click.
- $meta-creative-brief
- $ads-budget-reallocator
03 / Storefront + Performance
12-point theme audit covering Core Web Vitals, render-blocking resources, image optimization, mobile UX, and a11y. Plus a feed rebuilder for Meta and Google.
- $shopify-theme-audit
- $product-feed-rebuild
04 / Retention + Lifecycle
Diagnose your Klaviyo flows. Auto-respond to reviews in your brand voice. Catch the gaps in your post-purchase sequence.
- $klaviyo-flow-doctor
- $review-responder
05 / Operations + Catalog
Decide which SKUs to keep, cut, or feature. Pre-flight every launch with a 28-point checklist before you flip the lights on.
- $sku-rationalizer
- $launch-checklist
06 / Brand + Customer
Triage support tickets without losing your tone. Lint your copy against your brand voice file so every email sounds like the same brand.
- $support-ticket-triager
- $brand-voice-linter
// outcomes (from real implementations)
12
Skills shipping today
+4
Free as they ship
Live
Run by Michael in client work
Free
Future updates included
// common questions
What is a Claude Code skill?
A skill is a folder containing a single skill.md file with a system prompt and a defined procedure. You drop the folder into your project's .claude/skills/ directory and Claude Code can invoke the skill by name. It is the difference between copying a prompt out of a doc and running a command from your shell.
Do I need Claude Code Max or Claude Pro?
These skills run on any Claude Code tier. The agent orchestration patterns benefit from larger context windows on Max, but the core pack works on standard Claude.
How do I install the pack?
After purchase, your dashboard shows a Download Bundle button. Unzip into ~/.claude/skills/ for global access, or into .claude/skills/ inside any project for repo-local access. Each skill folder contains its own skill.md plus optional reference files.
What about the 4 future skills?
PDP conversion audit, post-purchase flow builder, returns and margin analyzer, and onsite popup audit. They ship free to anyone who already bought the pack. You get an email and a dashboard notification the day each one lands.
Can I customize the skills?
Yes. Every skill is a readable Markdown file with YAML frontmatter and a prompt body. Fork them, rename them, adjust them. The pack is a starting point.
Is this a glorified prompt library?
It is the opposite. Prompt libraries are prompts you copy-paste. Skills are complete workflows with system prompts, metadata, example invocations, and documented limitations. They run autonomously under Claude Code rather than getting pasted into a chat.
// proof: this is what it looks like in production
Q1 2026
A C-Suite Agent Framework That Drafts Memos in the Founder's Voice.
Pre-meeting prep dropped from hours to minutes. Decision documents arrive in the right voice on the first draft.
Q2 2024
How Server-Side Meta CAPI Recovered 30 to 40 Percent of Lost Conversions.
~35% more conversions tracked. Accurate ROAS for the first time in 18 months.
Not sure which tier? Start at the bottom. Work up.