FILE / FACTORY-13 / SHOPIFY OPS BREAKDOWN
13 AGENTS LIVE API · TOKEN-BILLED 2026 · TENTEN
CASE STUDY / SOLO-OPERATOR ECONOMICS

An agency.
One desk, thirteen robots,
a queue of dropshippers.

A solo operator in front of a wall-mounted LG monitor split into a 3×2 grid of six Claude Code windows. A second identical grid runs on a vertical display beside it. One more window on the MacBook within arm's reach. Thirteen agents simultaneously building Shopify stores — each in its own context, each owning a layer of the build. Six to seven finished stores per day. $800 each. Less than $80/day in API spend. No team. No managers. No support. Just him, the monitor, and the token counter ticking in every header.

13
Parallel Claude Agents
$800
Per Finished Store
~$80/day
Total API Spend, All 13
6–7/day
Stores Shipped, One Operator
CHAPTER01
A factory floor that fits on one desk.

The whole operation is one person, one chair, three displays. The wall-mounted ultrawide carries the first 3×2 grid of Claude Code instances. A second display, rotated vertical, holds another 3×2 grid. The MacBook on the desk runs the thirteenth. Wires hanging out, monitor bolted to a stand, no office.

He's not on a flat-rate plan — he's on API billing, metered by tokens. The arithmetic is the entire business model: thirteen parallel agents pay for themselves on the very first client of the day, because every finished store sells for $800 and the thirteen windows together burn less than $80 across a full day of work.

LG ULTRAWIDE — DISPLAY 01 ● REC
W01
Founder
Engineer
● ORCHESTRATING
W02
Catalog
Builder
● 80 SKUs
W03
Copy
Rewrite
● WRITING
W04
Homepage
Layout
● BUILDING
W05
Brand
System
● TOKENS
W06
Cart &
Checkout
● CONFIG
VERTICAL DISPLAY — DISPLAY 02 ● REC
W07
Email
Flows
● 30 CHAINS
W08
Banner
Designer
● 50 BANNERS
W09
Logo &
Identity
● ITERATING
W10
SEO &
Meta
● INDEXING
W11
Analytics
+ A/B
● WIRING
W12
Apps &
Plugins
● INSTALL
MACBOOK ● REC
W13
QA & Pre-launch
Auditor
● TESTING ORDER #4

Each window is a Claude Code session in its own working directory, with its own CLAUDE.md, its own MCP connections, and its own scope of files it's allowed to touch. They do not interfere with each other because they don't share a context window — the orchestration happens through the filesystem, the Shopify Admin API, and the operator's eyes.

You are my new founder-engineer.
— Window #1, system prompt, line 1
CHAPTER02
The system prompt that closed the assistant-or-employee debate.

Most operators talk to Claude like an intern — "can you help me with...", "what do you think about...". He skipped that. The first window's system prompt is one line that immediately re-levels the relationship:

# /CLAUDE.md — Window 01 (Orchestrator)

You are my new founder-engineer.

You don't hint. You don't advise. You don't supplement.
You own the result. The store ships, or it doesn't —
and that is on you.

# Working directory
/projects/client-{{client_id}}/

# Authority
- Read/write all files in this project.
- Call the Shopify Admin API via the MCP server already configured.
- Spawn sub-tasks in worktrees W02–W13 by writing briefs to /briefs/.
- Mark a brief done by moving it to /briefs/_done/.

# Definition of done
A live, working store at https://{{slug}}.myshopify.com with:
80 products, payment + shipping live, 30 email flows armed,
homepage + 50 banners shipped, GA4 + pixels firing,
QA report green. No half-finished stores leave the floor.

That single framing — founder-engineer, not assistant — does more for output quality than any prompt-engineering trick. It tells the model what posture to take when something is ambiguous: decide, document, move on. Don't stop and ask.

Why this matters: on a token-billed plan, every clarifying question is paid latency. Re-leveling the agent to "founder-engineer" cuts the back-and-forth by an order of magnitude because the model defaults to making the call and noting it in a decisions log, rather than pausing the workflow.
CHAPTER03
The thirteen agents, by layer of the store.

Every agent gets one slice of the store and a tight, deliverable-shaped scope. The orchestrator (W01) is the only one that talks to the operator. Everyone else reads a brief, executes, writes files, and signs off. Below: the full roster.

W01TIER · ORCHESTRATOR

The Founder-Engineer

Intake · Brief writing · Hand-off · Sign-off
Reads the client's intake form, generates the project skeleton, writes the briefs that go to W02–W13, and reviews the QA report. The only window that ever sees the operator's voice.
OUT › /briefs/*.md, /decisions.md, final delivery email
W02TIER · CATALOG

Catalog Builder

80 products · Variants · SKUs · Categories
Pulls supplier feed (CJ Dropshipping / AliExpress / Spocket via MCP), de-dupes, builds 80 products with variants, weights, prices, and tag taxonomy. Pushes via Shopify Admin API.
OUT › 80 products live in store, /catalog.json snapshot
W03TIER · COPY

Copy Rewrite

Product descriptions · PDPs · Microcopy
Rewrites every supplier description in brand voice. Generates feature/benefit blocks, FAQ sections, size guides where relevant, and trust microcopy on PDPs. Localizes if multi-region.
OUT › 80 PDP descriptions pushed back to W02's products
W04TIER · HOMEPAGE

Homepage Layout

Niche-specific homepage · Section composition
Reads the client's niche brief and chooses sections — hero, featured collections, social proof, value props, cross-sell rail. Builds via Shopify section.liquid templates or Shopify 2.0 JSON templates.
OUT › /sections/*.liquid + /templates/index.json
W05TIER · BRAND

Brand System

Tokens · Type · Color · Theme settings
Locks the design tokens — color palette, typography pairing, spacing rhythm, button radii — into settings_data.json so the rest of the store inherits them. Generates a one-page brand sheet for the client.
OUT › /config/settings_data.json + /brand-sheet.pdf
W06TIER · CHECKOUT

Cart & Checkout

Payments · Shipping zones · Tax · Cart UX
Wires Shopify Payments (or Stripe), configures shipping zones country-by-country, sets tax behavior, builds the slide-out cart with progress bars and free-shipping threshold logic.
OUT › Live checkout, all payment methods test-passed
W07TIER · LIFECYCLE

Email Flows

30 emails · Klaviyo / Shopify Email · Warming
Builds 30 lifecycle messages: welcome, browse-abandon, cart-abandon, post-purchase 1/2/3, winback 30/60/90, replenishment, VIP, broadcast templates. Loads them as Klaviyo flow JSON or Shopify Email campaigns.
OUT › 30 emails armed in Klaviyo, all triggers active
W08TIER · VISUAL

Banner Designer

50 banners · Hero / collection / promo / social
Generates 50 banners across the store, in the brand system locked by W05. Pulls product photography from W02, runs background-cleanup, composes hero + collection + promo + social sizes. Outputs WebP + AVIF.
OUT › /assets/banners/* — 50 files, WebP+AVIF
W09TIER · IDENTITY

Logo & Identity

Logo · Wordmark · Favicon · OG images
Iterates 6–10 logo concepts, picks one in line with the brand brief, exports the full identity pack — wordmark, monogram, favicon at every size, social avatars, OG cards for sharing.
OUT › /assets/identity/* — full pack, vector + raster
W10TIER · SEO

SEO & Meta

Titles · Descriptions · Schema · Sitemap · Robots
Writes meta titles + descriptions for every page, builds product/organization/breadcrumb JSON-LD schema, validates the sitemap, sets canonical URLs, fills robots.txt.liquid. Submits to Google Search Console.
OUT › All pages indexable, GSC verified, schema valid
W11TIER · DATA

Analytics + A/B

GA4 · Meta Pixel · TikTok · A/B framework
Installs GA4 with enhanced e-comm events, Meta Pixel + Conversions API, TikTok Pixel, wires consent banner to fire correctly. Sets up first A/B test on the homepage hero using Shopify experiments or GrowthBook.
OUT › All pixels firing, A/B test live with 50/50 split
W12TIER · APPS

Apps & Plugins

Reviews · Upsells · Bundles · GDPR · Trust
Installs the standard stack — Judge.me for reviews, ReConvert for post-purchase upsells, Shopify Bundles, GDPR/cookies banner — and configures each app to the niche. No bloat: only apps that earn their slot.
OUT › 5–8 apps installed and configured, all justified
W13TIER · QA

QA & Pre-launch Auditor

Test orders · Broken links · Lighthouse · Mobile
Places test orders in every shipping zone, walks every page on desktop + mobile, runs Lighthouse, checks broken links, validates that every email actually fires, signs off green or kicks the brief back to the responsible agent.
OUT › /qa-report.md — green, or back to W02–W12 with fixes
CHAPTER04
How a single store moves through the floor in under three hours.
01
Intake — operator only

New dropshipper drops a brief into the queue: niche, target country, supplier preference, a few reference stores. The operator pastes it into W01 (Orchestrator). That's the only manual typing of the day for this client.

~5 min · operator-driven
02
Brief generation — W01 fans out

The orchestrator scaffolds the project directory and writes 12 briefs into /briefs/, one per downstream window. Each brief contains: the client niche, the brand voice, the deliverables, the file paths, and the deadline.

~10 min · W01 only
03
Parallel build — W02 through W12

The operator opens each downstream window, types "read your brief and execute", and walks away. W02 (Catalog) and W05 (Brand System) run first because everything downstream depends on them. W03, W04, W06–W10 kick off as soon as their dependencies clear. W08 (Banners) waits on W02's product photos and W05's tokens.

~90 min · 11 agents in parallel
04
Pixel + analytics — W11

Once the homepage and PDPs are stable, W11 installs all pixels and wires the first A/B experiment. Runs late on purpose, because installing pixels before the layout settles produces noisy events.

~15 min · sequential after W04
05
QA pass — W13

W13 is the bouncer. It places live test orders in three shipping zones, runs a full Lighthouse pass, walks every link, fires every email trigger. Green report → W01 packages the handoff. Red report → routed back to the responsible window with a fix list.

~20 min · W13 only
06
Hand-off — operator + W01

W01 writes the delivery email — store URL, admin login transfer instructions, brand sheet attached, 30-day success checklist. Operator scans it, hits send. $800 invoice clears. Window 01 immediately starts pulling the next client from the queue.

~10 min · operator-confirmed
The operator's whole day is two decisions: which client gets the next finished store, and which client to take in next. Everything between intake and delivery is the agents.
CHAPTER05
The math is obscene, and that's the entire point.

Traditional Agency

$3,500per store
  • Build time~2 weeks
  • Headcount5 people
  • Calls / revisions3–5 rounds
  • Throughput1 store / 2 wks
  • Monthly capacity~2 stores
  • Op. cost / month~$30,000

The 13-Agent Factory

$800per store
  • Build time~2.5 hours
  • Headcount1 operator
  • Calls / revisions0 (intake brief only)
  • Throughput6–7 stores / day
  • Monthly capacity~180–200 stores
  • Op. cost / month~$2,400 (API)

The unit economics, line by line.

Revenue per store+ $800.00
API spend per store (≈ $80/day ÷ 6.5 stores)− $12.30
Shopify Partner / dev store fee (amortized)− $0.00
Asset gen (image API, fonts, stock if any)− $4.00
Operator overhead (allocated)− $25.00
Gross margin per store≈ $758.70

Six finished stores a day, six days a week, fifty weeks a year, and the math is what the math is. The point isn't that anyone should chase that number — it's that the cost structure of building Shopify stores has been re-priced. A $3,500 build is now a $12 token spend wrapped in an $800 deliverable.

Caveat that matters: "$80/day across all 13 windows" assumes mostly Sonnet-class usage with aggressive prompt caching, pre-loaded skills/SOPs, and tight context windows per agent. If every window sat in Opus-class long-context mode all day, that number would be 4–8× higher. The economics still work — but the operator's discipline on which agent runs on which model is what makes the day cost $80 instead of $400.
CHAPTER06
The replication playbook, step by step.

Phase 1 — Workstation

Hardware & layout

  • One ultrawide or 4K display — at least 38" — set up to host a 3×2 tiling gridmacOS Stage Manager + Rectangle, or Windows FancyZones
  • One vertical secondary display — 27" rotated portrait works fine — second 3×2 grid
  • One laptop in arm's reach — for the 13th window, usually the QA/auditor agent
  • Tiling shortcuts memorized — you'll send 13 windows to specific cells dozens of times a dayMap Cmd+Opt+1..6 for the 6 cells of each grid

Phase 2 — Project skeleton

# Create the per-client template you'll clone for every new order.
mkdir -p ~/factory/_template
cd ~/factory/_template

# Top-level layout
mkdir -p briefs briefs/_done decisions assets/identity assets/banners \
         catalog config sections templates qa

# Per-window CLAUDE.md files — one per agent
for n in $(seq -w 01 13); do
  mkdir -p windows/W$n
  touch    windows/W$n/CLAUDE.md
done

# Shared MCP config
mkdir -p .mcp
touch    .mcp/shopify.json .mcp/klaviyo.json .mcp/supplier.json

Every new client is cp -R _template client-{slug}. The factory has a single source of truth for window prompts, file paths, and MCP wiring — touch it once, propagate to all future stores.

Phase 3 — The thirteen system prompts

Below are the three load-bearing prompts. Build the rest in the same shape — one role, tight scope, deliverable-shaped definition-of-done. (Full set of 13 lives in the template repo.)

W01 — Orchestrator (Founder-Engineer)

You are my new founder-engineer. You don't hint. You don't advise. You own the result.

# Working directory  /factory/client-{{slug}}/
# Authority          Read/write all files. Call Shopify Admin via MCP.

# Your only job
1. Read /intake.md (the client brief).
2. Decide niche, brand voice, regions, payment, shipping.
   Write decisions to /decisions/01-foundation.md.
3. Generate 12 briefs in /briefs/W02..W13.md. Each brief contains:
   - GOAL (one sentence)
   - INPUTS (paths or URLs to read)
   - OUTPUTS (exact file paths the agent must produce)
   - DEFINITION OF DONE (the test that must pass)
4. Wait. When all 12 briefs are in /briefs/_done/, run the
   final QA review and write /delivery.md (client email + handoff).

Never ask me a clarifying question. Make the call, log it, move.

W02 — Catalog Builder

You build the product catalog. 80 products. No more, no fewer.

# Read   /briefs/W02.md, /decisions/01-foundation.md
# Tools  MCP: supplier (CJ/AliExpress/Spocket), Shopify Admin
# Write  /catalog/products.json, then push to Shopify via API

# Quality bar
- 80 SKUs, deduped by image-hash + title-similarity.
- Price = supplier_cost × markup_table[niche].
- Variants normalized (size XS–XXL, color names canonicalized).
- Tags from the niche taxonomy in /decisions/01-foundation.md.
- Every product must have ≥3 images and a non-empty body_html.
  Body_html is a placeholder — W03 (Copy Rewrite) will replace it.

Done = 80 products visible in Shopify Admin AND
       /catalog/products.json snapshot committed.

W13 — QA & Pre-launch Auditor

You are the bouncer. No half-finished store ships.

# Read   /briefs/W13.md, all /briefs/_done/*.md
# Tools  MCP: Shopify Admin, headless browser (Playwright), Lighthouse

# Run, in order
1. Test orders in 3 shipping zones (default + 2 international).
   Verify each completes through to confirmation page.
2. Walk every page (homepage, all collections, 5 random PDPs,
   cart, checkout, account, search, 404, policy pages) on
   desktop + mobile breakpoints.
3. Lighthouse on homepage + 1 PDP. Performance > 70 mobile.
4. Verify every email flow fires by triggering each event.
5. Validate JSON-LD schema, sitemap, robots, canonical tags.

# Output
/qa/report.md with: pass/fail per check, screenshots of failures,
and — if any fail — a /qa/fixes-for-W{nn}.md addressed to the
responsible window. Do not mark done if any check is red.

Phase 4 — MCP & tool wiring

SHOPIFY MCP

Use the official Shopify Dev MCP server or a custom one wrapping the Admin GraphQL API. W02, W03, W04, W06, W10, W11, W12, W13 all need it. Scope tokens by store — never one token across all clients.

SUPPLIER MCP

CJ Dropshipping, Spocket, or AliExpress via a wrapper. W02 needs it. Cache the feed locally once per day — supplier APIs are rate-limited and slow, and you do not want every catalog agent re-pulling the same SKUs.

KLAVIYO MCP

Custom server around Klaviyo's REST API. W07 needs it to push 30 lifecycle messages. Build this once, version-control the flow JSON templates per niche.

IMAGE / DESIGN APIS

For W08 (banners) and W09 (logo), wire image generation (Higgsfield / Nano Banana 2 / Soul) and a background-removal service. W08's 50-banner pass is the single biggest non-Anthropic line item in the daily cost.

Phase 5 — Operating rhythm

The operator's day, hour by hour

  • 09:00 — Pull the queue. Pick today's 6 clients in priority order.Triage: paid-in-full first, repeat clients second, niche-fit third.
  • 09:15 — Kick off Client #1. Paste intake into W01. W01 writes briefs.~10 min before W02–W12 are loaded.
  • 09:30 — Open W02–W12 for Client #1. Type "execute brief". Walk away.
  • 10:00 — Kick off Client #2 in a parallel directory.Yes — multiple clients can run on the same 13 windows if each agent is given a distinct working directory per turn. Discipline matters here.
  • 11:30 — Client #1 hits W13 (QA). Operator scans report.
  • 12:00 — Client #1 ships. W01 sends delivery email. Cash clears.
  • 12:00 → 18:00 — Same loop, staggered, 5 more times.
  • End of day — Archive completed projects. Commit decisions logs. Refill queue.
CHAPTER07
Skills, SOPs, and the files that make it repeatable.

The reason store #200 ships as cleanly as store #1 is that every recurring decision has been promoted into a Skill. Skills are the durable knowledge layer — markdown files Claude Code loads automatically based on description match. Below: the minimum viable Skill set the factory runs.

SKILL · 01CATEGORY · CATALOG

shopify-product-loader

For W02 — bulk product creation via GraphQL Admin API
Triggers when an agent says "load 80 products into Shopify". Encodes the GraphQL mutations, rate-limit handling, image upload retries, variant/option construction. Removes 90% of the "how do I call the Shopify API" tokens.
FILE › /skills/shopify-product-loader/SKILL.md
SKILL · 02CATEGORY · COPY

pdp-copywriter-{niche}

For W03 — niche-specific PDP voice templates
One Skill per major niche (beauty, pet, home, fitness, kitchen). Each contains the voice guide, the section structure (hook → benefits → features → trust → FAQ), and 3 reference examples. Operator builds one new niche Skill per ~5 stores in unfamiliar territory.
FILE › /skills/pdp-copywriter-{niche}/SKILL.md
SKILL · 03CATEGORY · BRAND

tokens-to-shopify-settings

For W05 — design tokens → settings_data.json mapping
Encodes the exact mapping between W05's design-token output and the Shopify theme's settings_data.json structure. Without this Skill, W05 spends tokens re-discovering what setting key controls which CSS variable.
FILE › /skills/tokens-to-shopify-settings/SKILL.md
SKILL · 04CATEGORY · LIFECYCLE

klaviyo-flow-pack-30

For W07 — the 30-email canonical structure
The 30 lifecycle messages pre-templated. Each variant has triggers, delays, conditional splits, and brand-voice slots. W07 just fills in the brand voice and presses go.
FILE › /skills/klaviyo-flow-pack-30/SKILL.md + 30 .json templates
SKILL · 05CATEGORY · VISUAL

banner-grid-50

For W08 — the 50-banner sizing matrix
Sizes, aspect ratios, file naming, output formats (WebP+AVIF), and where each banner gets placed in the theme. W08 follows the matrix and only spends creative tokens on composition, not specs.
FILE › /skills/banner-grid-50/SKILL.md
SKILL · 06CATEGORY · DATA

pixels-and-consent

For W11 — the canonical pixel + consent install
GA4 + Meta CAPI + TikTok pixel + consent banner, wired so events only fire after explicit consent. The pixel order, the dataLayer schema, the validation steps — all in one Skill so W11 produces the same install every time.
FILE › /skills/pixels-and-consent/SKILL.md
SKILL · 07CATEGORY · QA

prelaunch-qa-checklist

For W13 — the audit script that gates shipping
The full Lighthouse + test-order + email-trigger walk. Drives W13 deterministically so the same checks run on every store. The single most important Skill — without it, QA drifts and stores ship with bugs.
FILE › /skills/prelaunch-qa-checklist/SKILL.md
SKILL · 08CATEGORY · ORCHESTRATION

brief-writer

For W01 — converts intake into 12 downstream briefs
The Skill that turns a 200-word client intake into 12 deliverable-shaped briefs. Encodes the brief schema (GOAL / INPUTS / OUTPUTS / DEFINITION OF DONE) and the dependency graph between agents.
FILE › /skills/brief-writer/SKILL.md

SKILL.md template — the shape every Skill takes

---
name: shopify-product-loader
description: Use when bulk-creating Shopify products via the Admin GraphQL API.
  Triggers on phrases like "load 80 products", "push catalog to Shopify",
  "bulk product create". Handles variants, images, tags, metafields,
  rate-limiting, and retries. Required for W02 (Catalog Builder).
---

# When to use
- W02 building a fresh catalog from supplier feed
- Any agent that needs to write products to Shopify

# Inputs you must have
- A products.json with the canonical schema (see below)
- A valid Shopify Admin API token in env var SHOPIFY_ADMIN_TOKEN

# Steps
1. Validate products.json against schema in /skills/.../schema.json
2. Chunk products in batches of 25 for GraphQL `productSet` mutation
3. Call mutation; on rate-limit (429), exponential backoff
4. For each successful product, upload images via stagedUploads
5. Verify each product is visible in Admin (GET productByHandle)

# Output
- All 80 products live in store
- /catalog/_loaded.json with Shopify product GIDs

# Failure handling
- If < 80 products land: write /catalog/_failed.json with reason per SKU
- Never mark "done" with partial load. Hand back to W01.
The Skills are the moat. The 13 system prompts are public — anyone can copy them. What's hard to copy is 8 weeks of accumulated Skills that encode every weird edge case you hit on stores #1 through #50. Store #200 ships in 2.5 hours because the Skills already know what tripped you up on store #37.
CHAPTER08
What could break this, and what doesn't fit on the slide.

The story is real-shaped, and the architecture is reproducible. A few caveats worth being honest about before anyone fires a credit card at the API.

1 · "$80/DAY" IS A FLOOR, NOT A CEILING

Heavy banner generation, Opus on every window, long context windows, no prompt caching — and you can run $400/day on the same setup. The $80 number assumes ruthless model routing (cheap models for boilerplate, expensive only when needed) and aggressive caching.

2 · QUALITY DRIFT IS REAL

At 6 stores/day, store #200 will look very similar to store #5 unless niches differ enough. Dropshippers paying $800 don't always notice; dropshippers building lasting brands will. The model only ships what the QA agent (W13) and the operator's eye let through.

3 · LIABILITY IS UNCAPPED

Auto-built stores with auto-generated copy can ship trademark violations, regulated-product mistakes (supplements, cosmetics, electronics), or false claims. W13's QA needs a compliance check, especially if operating across regions with different ad/product law.

4 · SUPPLIER FRAGILITY

All 13 agents collapse if the supplier feed dies, the Shopify Admin API rate-limits, or Klaviyo throttles new accounts. Build queue + retry into every MCP server, and have a local cache so a 30-min outage doesn't kill the day's throughput.

5 · CLIENT EXPECTATION MANAGEMENT

$800 buys a great launch package, not a custom design system. The intake brief has to make that obvious. The operators who scale this without burnout sell "a working store, ready Monday", not "your forever brand."

6 · IT IS NOT TRULY HANDS-OFF

"He decides 2 things" is the marketing line. The reality: the operator scans every QA report, fields edge-case escalations from W01, handles supplier outages, and answers client emails. Less than 5 hours of attention per day — but not zero.


What this means for everyone else.

Forget the $4,800/day headline. The architectural lesson is the one that travels: thirteen narrow agents, each owning one layer, coordinated through the filesystem, with one human deciding the gates — that pattern works for any production line where the deliverable can be decomposed into independent layers. Shopify stores today. Real estate listing sites tomorrow. SaaS landing pages the day after.

The "founder-engineer" framing isn't even Shopify-specific. You don't hint, you don't advise, you own the result is a posture-setting move that re-prices any agent's output, in any domain, by removing the assistant-shaped politeness tax from every turn. Try it once on a system prompt for any agent you currently call a "helper" — and watch what changes.

The headline is the wrong takeaway. The right takeaway: a single operator, with the right thirteen prompts and eight Skills, has just demonstrated that the cost basis of a Shopify store build is now closer to lunch than to two weeks of agency time. Everyone selling at the old price has a window to either match this, or sell something else.