All skills
rshankras avatar

/store-growth-audit

@32566aa

Stage-by-stage audit of an app's App Store growth machinery against a 54-item P0–P9 playbook — every item scored from an App Store Connect MCP call, a codebase check, or an explicit question to the user, then routed to the skill or command that fixes it. Read-only on App Store Connect. Use for a growth audit or scorecard, a pre-launch growth plan, a quarterly re-audit, or "which growth levers am I missing."

Use this Skill: https://skilld.dev/gh/rshankras/claude-code-apple-skills/store-growth-audit

This session only. Nothing lands on disk.

detection-playbook.md

≈1.6k tokens on demand. Your agent reads this file only when SKILL.md points to it.

Detection Playbook — gather once, evaluate 54

The audit's evidence-gathering machinery. Run all three passes up front, store the evidence under the keys below, then evaluate every checklist item against the stored evidence. Never interleave gathering with scoring; never ask MANUAL questions one at a time.

Phase-scoped runs (SKILL.md → Scoped Runs) execute only the table rows whose "Feeds items" column intersects the scope — same three-pass order, smaller batches.

Pass 1 — MCP evidence batch (read-only, ~14 calls)

One call per row; each populates an evidence key consumed by the listed items.

Call Evidence key Feeds items
list_apps EV.app (appId, state, live?) mode detection
get_metadata EV.meta (name, subtitle, keywords, description, promo text, categories, URLs, version, what's-new) P1.1 P1.2 P1.6 P1.7 P2.2 P2.3 P2.4 P6.3 P9.3
list_locales EV.locales (localized locales + per-locale fields) P1.2 P3.1 P3.4
get_app_pricing EV.pricing (base price + per-storefront overrides) P3.3
list_price_points EV.pricepoints (available tiers per storefront) P3.3
get_availability EV.territories (territories, platforms) P3.2 P6.4 P7.6
list_iap EV.iaps (IAPs, display names, familySharable) P1.4 P7.7 P8.7
list_subscription_groups + list_subscriptions EV.subs (groups, tiers, offers, familySharable) P0.2 P8.1 P8.2 P8.7
list_app_events EV.events (events, badges, dates, metadata) P1.5 P5.4
list_experiments EV.ppo (PPO experiments + states + dates) P2.1 P5.1 P9.4
list_custom_pages EV.cpp (custom product pages) P5.2 P7.1
get_analytics_report EV.analytics (impressions, page views, CVR, retention, sources) P0.3 P3.4 baseline snapshot
get_sales_report EV.sales (units/proceeds, by territory) P3.4 baseline snapshot
list_reviews EV.reviews (recent reviews, stars, developer-response present?) P4.2

Pre-launch mode: skip the batch (no live app); EV.meta-class evidence comes from .planning/ASO.md, APP.md, ROADMAP.md instead. Analytics not configured yet → note "baseline lands next cycle — route to store-signals to set up" (never call setup from the audit).

Pass 2 — codebase grep batch

One pass over the app repo (skip if no codebase is at hand — affected items fall back to MANUAL).

Pattern (Grep/Glob) Evidence key Feeds items
requestReview|SKStoreReviewController|AppStore.requestReview EV.code.reviewprompt P4.1
setAlternateIconName|CFBundleAlternateIcons + alternate entries in *.xcassets EV.code.alticons P2.5
import AppIntents|AppShortcutsProvider|INIntent EV.code.intents P6.5
CSSearchableItem|CoreSpotlight EV.code.spotlight P6.5
import WidgetKit / import ActivityKit EV.code.widgets / EV.code.liveactivity P9.1
UNUserNotificationCenter|requestAuthorization EV.code.notifications P9.1
onboarding flow markers (Onboarding, first-run flags) EV.code.onboarding P9.1
com.apple.developer.storekit.external-purchase in *.entitlements EV.code.extpurchase P8.4
*.storekit config, SubscriptionStoreView|StoreKit paywall views EV.code.paywall P8.1
deletion flow (deleteAccount, account-deletion UI) EV.code.accountdeletion P2.4
*.lproj dirs / locales inside *.xcstrings EV.code.locales P3.4
platform targets in project.pbxproj (SDKROOT, SUPPORTED_PLATFORMS, Catalyst flag) EV.code.platforms P6.4
fastlane/ASC-API scripts (fastlane/, deliver, app_store_connect_api_key, metadata upload scripts) EV.code.automation P3.5
new-OS API adoption (imports gated by current-year #available) EV.code.newapis P6.2

Pass 3 — the MANUAL question batch

One AskUserQuestion round covering every MANUAL/HYBRID item the first two passes can't settle. Ask only what's still unknown; phrase options so each maps directly to a status. "Not sure" is always an option and always scores 🟠 unverified.

# Question Maps to
1 Enrolled in the App Store Small Business Program? (enrolled / not enrolled / ineligible >$1M / not sure) P0.1
2 Billing Grace Period AND Billing Retry enabled in ASC? (both / one / neither / no subs) P0.2
3 Checked ASC peer-group benchmarks for this app? (yes, recently / long ago / never) P0.3
4 Apple Ads: any discovery campaign run for keyword research? Exact-match campaigns? Brand term defended? P0.4 P7.1 P7.2
5 Is the developer/brand name a deliberate, descriptive choice? P1.3
6 Privacy nutrition labels reviewed against current data collection? P2.4
7 Do you release phased + manual (not auto-release on approval)? P4.3
8 Any featuring nomination submitted in the last 6 months? A calendar for them? P6.1
9 Landing page live? Smart App Banner on it? Ever listed on deal sites during a price drop? P6.6 P7.5
10 Launch-spike playbook used at last major release (Product Hunt + press + newsletter inside 48h)? P7.4
11 Pre-orders ever used (incl. regionally for new storefronts)? Offer codes distributed? TestFlight public link used as a waitlist? P7.6 P7.7 P7.8
12 Paywall/pricing experiments run in the last two quarters? P8.1
13 US web-checkout / External Purchase Link entitlement: shipped, planned, or not considered? P8.4
14 Any cross-developer bundle conversations? Institutional (school/business) buyers in your audience? P8.5 P8.6
15 ASA search-term report folded into keywords in the last quarter? P9.2

Trim the batch: drop questions already answered by evidence (e.g. EV.code.extpurchase present → ask only about the commission-flip architecture, not whether it shipped). In portfolio mode ask only the account-level questions (1, 3, 4-brand, 8) once for all apps.

Freshness caveats

  • Analytics/sales reports arrive on Apple's schedule; a missing report is "no data yet", not 🔴.
  • EV.reviews developer-response coverage is the signal for P4.2 — count unanswered 1–3★ reviews from the last 90 days, don't judge tone here (that's review-response-writer's job).
  • Evidence strings must be terse and dated where MANUAL: SBP: enrolled (user, 2026-07), locales: en-US, de, ja, PPO: last experiment 2026-03 (stale).

Source: SKILL.md on GitHub

No alerts2mo3 checks · Risk SAFE
  • Gen Agent Trust Hub2mo

    The skill is a comprehensive growth audit utility for App Store applications that evaluates app machinery against a 54-item playbook. It uses specialized MCP tools to access sensitive App Store metadata, sales reports, and analytics, and performs local codebase inspections using grep. The primary security consideration identified is an indirect prompt injection surface created by the ingestion of untrusted user-generated review text from the App Store. The skill operates in a read-only capacity for App Store Connect and writes audit results to a local file.

  • Socket2mo

    No alerts

  • Snyk2mo

    Risk: LOW · No issues

Signed by skilld at 32566aa. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 months ago.

Steadyupdated 3 months ago
What it can do
Reads files Edits files
MCP servers
asc-metadata
last_verified
2026-07-16
review_by
2027-06-22
All 20 allowed tools
ReadWriteGlobGrepAskUserQuestionmcp__asc-metadata__list_appsmcp__asc-metadata__get_metadatamcp__asc-metadata__list_localesmcp__asc-metadata__get_app_pricingmcp__asc-metadata__list_price_pointsmcp__asc-metadata__get_availabilitymcp__asc-metadata__list_iapmcp__asc-metadata__list_subscription_groupsmcp__asc-metadata__list_subscriptionsmcp__asc-metadata__list_app_eventsmcp__asc-metadata__list_experimentsmcp__asc-metadata__list_custom_pagesmcp__asc-metadata__get_analytics_reportmcp__asc-metadata__get_sales_reportmcp__asc-metadata__list_reviews

README badge

README badge for rshankras/claude-code-apple-skills/store-growth-audit