cr1707/shopify-audit-agents
Audit any online store with a team of Claude Code agents: 8 area leads, 24 specialists, a verifier and an editor. Evidence-only findings and copy-paste fixes for Claude.
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code $ npx degit cr1707/shopify-audit-agents/agents ./rig-shopify-audit-agents/agents $ npx degit cr1707/shopify-audit-agents/skills ./rig-shopify-audit-agents/skills
MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit cr1707/shopify-audit-agents/skills/audit .claude/skills/audit $ curl -fsSL --create-dirs -o .claude/agents/copy-consistency.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/copy-consistency.md $ curl -fsSL --create-dirs -o .claude/agents/copy-quality.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/copy-quality.md $ curl -fsSL --create-dirs -o .claude/agents/copy-value-prop.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/copy-value-prop.md $ curl -fsSL --create-dirs -o .claude/agents/design-accessibility.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/design-accessibility.md $ curl -fsSL --create-dirs -o .claude/agents/design-consistency.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/design-consistency.md $ curl -fsSL --create-dirs -o .claude/agents/design-imagery.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/design-imagery.md $ curl -fsSL --create-dirs -o .claude/agents/editor.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/editor.md $ curl -fsSL --create-dirs -o .claude/agents/growth-ads.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/growth-ads.md $ curl -fsSL --create-dirs -o .claude/agents/growth-consent.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/growth-consent.md $ curl -fsSL --create-dirs -o .claude/agents/growth-tracking.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/growth-tracking.md $ curl -fsSL --create-dirs -o .claude/agents/lead-copy.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/lead-copy.md $ curl -fsSL --create-dirs -o .claude/agents/lead-design.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/lead-design.md $ curl -fsSL --create-dirs -o .claude/agents/lead-growth.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/lead-growth.md $ curl -fsSL --create-dirs -o .claude/agents/lead-products.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/lead-products.md $ curl -fsSL --create-dirs -o .claude/agents/lead-qa.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/lead-qa.md $ curl -fsSL --create-dirs -o .claude/agents/lead-tech.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/lead-tech.md $ curl -fsSL --create-dirs -o .claude/agents/lead-trust.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/lead-trust.md $ curl -fsSL --create-dirs -o .claude/agents/lead-ux.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/lead-ux.md $ curl -fsSL --create-dirs -o .claude/agents/products-catalogue.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/products-catalogue.md $ curl -fsSL --create-dirs -o .claude/agents/products-merchandising.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/products-merchandising.md $ curl -fsSL --create-dirs -o .claude/agents/products-pricing.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/products-pricing.md $ curl -fsSL --create-dirs -o .claude/agents/qa-breakage.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/qa-breakage.md $ curl -fsSL --create-dirs -o .claude/agents/qa-buy-buttons.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/qa-buy-buttons.md $ curl -fsSL --create-dirs -o .claude/agents/qa-mobile-layout.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/qa-mobile-layout.md $ curl -fsSL --create-dirs -o .claude/agents/tech-seo.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/tech-seo.md $ curl -fsSL --create-dirs -o .claude/agents/tech-speed.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/tech-speed.md $ curl -fsSL --create-dirs -o .claude/agents/tech-stack.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/tech-stack.md $ curl -fsSL --create-dirs -o .claude/agents/trust-policies.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/trust-policies.md $ curl -fsSL --create-dirs -o .claude/agents/trust-proof.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/trust-proof.md $ curl -fsSL --create-dirs -o .claude/agents/trust-retention.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/trust-retention.md $ curl -fsSL --create-dirs -o .claude/agents/ux-first-screen.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/ux-first-screen.md $ curl -fsSL --create-dirs -o .claude/agents/ux-friction.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/ux-friction.md $ curl -fsSL --create-dirs -o .claude/agents/ux-purchase-path.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/ux-purchase-path.md $ curl -fsSL --create-dirs -o .claude/agents/verifier.md https://raw.githubusercontent.com/cr1707/shopify-audit-agents/main/agents/verifier.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 16.2k rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(./.env.*)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(git push --force:*)",
"Read(**/*.pem)",
"Read(.env)",
"Read(~/.aws/**)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Bash(rm -rf *)",
"Bash(sudo *)",
"Bash(sudo:*)",
"Bash(rm -rf /)",
"Bash(git push -f:*)",
"Read(.env.*)",
"Bash(git reset --hard*)",
"Bash(rm -rf /*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(gh pr merge *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(gh pr merge:*)",
"Bash(gh pr create *)",
"Bash(git reset --hard:*)",
"Bash(npm publish:*)",
"Bash(git rebase *)"
]
}
} Skills (1)
Subagents (34)
| copy-consistency model: sonnet | Copy and messaging specialist. Finds contradictions between pages: shipping times and costs, return windows, trial lengths, guarantees, prices and counts. Writes one findings file. Used by lead-copy. |
| copy-quality model: sonnet | Copy and messaging specialist. Checks correctness and clarity: spelling, grammar, tone, jargon, language mix, and whether product descriptions answer buyer questions. Writes one findings file. Used by |
| copy-value-prop model: sonnet | Copy and messaging specialist. Reads the value proposition and calls to action: what the store says it sells, to whom and why it is better. Writes one findings file. Used by lead-copy. |
| design-accessibility model: sonnet | Design and accessibility specialist. Summarises accessibility problems found by axe-core plus contrast, text size, tap targets, alt text, labels and language. Writes one findings file. Used by lead-de |
| design-consistency model: sonnet | Design and accessibility specialist. Reviews visual hierarchy and consistency: typography, colour, buttons, spacing and leftovers from the theme demo. Writes one findings file. Used by lead-design. |
| design-imagery model: sonnet | Design and accessibility specialist. Reviews product and lifestyle imagery: coverage per product, consistency across the grid, quality, and sharing images. Writes one findings file. Used by lead-desig |
| editor model: inherit | Editor-in-chief of the audit. Merges verified findings from all areas, ranks them by business impact, writes the executive summary and the final findings.json, then renders the report, the fix-with-Cl |
| growth-ads model: sonnet | Tracking, ads and privacy specialist. Checks acquisition hygiene: public ad libraries, ad landing pages, social links, and product data for Shopping feeds. Writes one findings file. Used by lead-growt |
| growth-consent model: sonnet | Tracking, ads and privacy specialist. Describes cookie-consent behaviour factually: what is shown, how easy declining is, and what still fires before a choice and after Decline. Writes one findings fi |
| growth-tracking model: sonnet | Tracking, ads and privacy specialist. Checks measurement: which analytics and ad pixels load and send data, duplicates, events around Add to cart, and campaign link hygiene. Writes one findings file. |
| lead-copy model: sonnet | Lead auditor for Copy and messaging (what the words say: the value proposition, contradictions between pages, clarity and correctness). In deep mode runs its three specialists (copy-value-prop, copy-c |
| lead-design model: sonnet | Lead auditor for Design and accessibility (visual quality, consistency, imagery and accessibility). In deep mode runs its three specialists (design-consistency, design-imagery, design-accessibility) i |
| lead-growth model: sonnet | Lead auditor for Tracking, ads and privacy (measurement, consent behaviour, ads and acquisition hygiene). In deep mode runs its three specialists (growth-tracking, growth-consent, growth-ads) in paral |
| lead-products model: sonnet | Lead auditor for Products, offer and pricing (the catalogue, prices, discounts, bundles and merchandising). In deep mode runs its three specialists (products-catalogue, products-pricing, products-merc |
| lead-qa model: sonnet | Lead auditor for Bugs and broken things (things that are broken: buttons that add the wrong product, dead links, errors, layouts that break on phone). In deep mode runs its three specialists (qa-buy-b |
| lead-tech model: sonnet | Lead auditor for Tech, speed and SEO (speed and Core Web Vitals, technical SEO, apps and scripts (the backend a visitor feels)). In deep mode runs its three specialists (tech-speed, tech-seo, tech-sta |
| lead-trust model: sonnet | Lead auditor for Trust, care and retention (policies, social proof, customer support and what brings buyers back). In deep mode runs its three specialists (trust-policies, trust-proof, trust-retention |
| lead-ux model: sonnet | Lead auditor for UX and conversion (how easily a visitor understands the offer and gets from landing to cart). In deep mode runs its three specialists (ux-first-screen, ux-purchase-path, ux-friction) |
| products-catalogue model: sonnet | Products, offer and pricing specialist. Audits the catalogue data: availability, images, descriptions, SKUs, duplicates, product types, freshness. Writes one findings file. Used by lead-products. |
| products-merchandising model: sonnet | Products, offer and pricing specialist. Reviews how products are presented and sold: home page selection, collections, cross-sell, bundles, badges. Writes one findings file. Used by lead-products. |
| products-pricing model: sonnet | Products, offer and pricing specialist. Audits prices and offers: permanent sales, compare-at prices, discount stacking signals, bundle maths, currency and free-shipping thresholds. Writes one finding |
| qa-breakage model: sonnet | Bugs and broken things specialist. Finds dead links, soft 404s, JavaScript errors, failed requests, broken images, leftover theme placeholder text and empty collections. Writes one findings file. Used |
| qa-buy-buttons model: sonnet | Bugs and broken things specialist. Clicks every add-to-cart button as a shopper and checks the right product reaches the cart; also sold-out and variant problems. Writes one findings file. Used by lea |
| qa-mobile-layout model: sonnet | Bugs and broken things specialist. Looks for layouts that break on a phone: sideways scrolling, overlapping or cut-off elements, text too small to read, bars that eat the screen. Writes one findings f |
| tech-seo model: sonnet | Tech, speed and SEO specialist. Checks technical SEO: titles, descriptions, headings, canonicals, structured data, sitemap coverage, robots rules and social tags. Writes one findings file. Used by lea |
| tech-speed model: sonnet | Tech, speed and SEO specialist. Reads PageSpeed Insights (lab and real-user field data), page weight and third-party cost; explains the business effect. Writes one findings file. Used by lead-tech. |
| tech-stack model: sonnet | Tech, speed and SEO specialist. Maps the app and script stack: overlapping apps, duplicate tags, third-party weight, and apps whose costs grow with traffic. Writes one findings file. Used by lead-tech |
| trust-policies model: sonnet | Trust, care and retention specialist. Checks policies and business identity: refund, shipping, terms, privacy, contact details, and how easy they are to find. Writes one findings file. Used by lead-tr |
| trust-proof model: sonnet | Trust, care and retention specialist. Reviews social proof and reassurance: reviews and ratings, where they sit, guarantees, badges and the brand story. Writes one findings file. Used by lead-trust. |
| trust-retention model: sonnet | Trust, care and retention specialist. Reviews what brings buyers back: email and SMS capture, support channels, loyalty and referral, and a manual test plan for lifecycle emails. Writes one findings f |
| ux-first-screen model: sonnet | UX and conversion specialist. Judges what a visitor sees and understands in the first screen, on phone and desktop, for the home page and product pages. Writes one findings file. Used by lead-ux. |
| ux-friction model: sonnet | UX and conversion specialist. Finds interruptions: pop-ups, stacked overlays, chat widgets, sticky bars, autoplay, and offers that open before the product reaches the cart. Writes one findings file. U |
| ux-purchase-path model: sonnet | UX and conversion specialist. Walks the path from landing to cart: navigation, collection pages, product page structure, where key facts sit, and the cart. Writes one findings file. Used by lead-ux. |
| verifier model: inherit | Fact-checker for audit findings. Re-opens the evidence for every candidate finding, re-runs live checks for serious claims, and confirms, downgrades or rejects each one. Run after the area leads, befo |
Similar rigs
alegauss/polyweave
A Claude Code plugin that makes a game's parts by declaring what each has to satisfy and searching for the parameters that do: 3D assets, sound loops and names, on Blender and Godot.
YOLO Cowboy 6.4k tok · · Q3
alegauss/claude-tray
Native Windows tray monitor for Claude Code usage — shows your rate-limit %, projects when you'll run out, and turns the icon red before you hit the limit. Plus a local 24h usage breakdown.
Pragmatist 6.9k tok · · Q15
alegauss/roadkeep
Keeps a project's roadmap, changelog, improvements and strategy files to one format - a schema at the point of insertion, not a linter after the fact. Python CLI + Claude Code plugin.
Automator 8.6k tok · · Q15
michaelmarconi/agentic-engineering
A software development lifecycle for Claude Code: specs, decision records and plans, independently reviewed, taking work from idea to verified code.
Orchestrator 7.7k tok · · Q5