agentleFS
Sign inStart free
datnpqdatnpq/gskill/skills/growth-audit/SKILL.md

growth-audit

Skill0 starsChanged 48 days agoMITView on GitHub

Audit and verify the whole Google marketing chain for a site — tags live on the page, GA4 key events and data quality, Search Console coverage, Ads conversion actions, product links (GA4↔Ads, GA4↔Search Console, Ads↔Search Console), billing, and a 48-hour post-launch check. Use for "check everything", "kiểm tra toàn bộ", "is my tracking working", or before/after launching a campaign.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. fliqrlabs/instagram-creator-skills★ 5same name
    2. super0510/GrowthLint★ 2same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository7 other files it gives its agents.

The file

Create a free account to mount
Files
SKILL.md

Growth audit

Work through the four layers. For each finding record: what, evidence, impact, fix. Fix what is safe to fix (code, links); report what needs the user.

Layer 1 — the live site

python ${CLAUDE_PLUGIN_ROOT}/scripts/audit_site.py https://example.com \
  --private /account /checkout --expect G-XXXXXXX AW-XXXXXXXXX

Checks: status codes, gtag loader and config IDs, url_passthrough, whether page_location strips query strings wholesale (gclid loss), robots.txt rules and sitemap line, sitemap validity and URL count, canonical, noindex on private paths, JSON-LD types. Then run check_gtag.js in a real browser with ?gclid=TEST&utm_source=google&x=secret.

Layer 2 — GA4

  • Home → are users/events arriving? Realtime shows your own visit?
  • Reports → Pages and screens: any personal data in titles? (names, emails) → fix now.
  • Traffic acquisition: localhost:* or staging referrals = test traffic mixed in; add a data filter / unwanted referral, and keep test environments on a different measurement ID.
  • Admin → Events → key events: purchase starred; others starred once they appear.
  • Admin → Product links: Google Ads links (Completed), Search Console links (present).

Layer 3 — Search Console

  • Sitemaps: Success, discovered ≈ <loc> count, last read recent.
  • Pages: indexed vs not; read reasons with search-console-seo §2 (young domains are mostly "Discovered – not indexed"; request indexing for key pages).
  • Settings → Associations: GA4 and Google Ads listed.

Layer 4 — Google Ads

  • Goals → Conversions → Summary → "View all conversion actions": each action's source, status, primary/secondary, count, window. "Pending" < 48 h after creation is normal.
  • Open each web action → confirm the trigger (event import / URL rule) matches a page that real users reach.
  • Data manager: Google tag present; GA4 linked; Search Console linked.
  • Campaigns: status column ("Eligible (Learning)" = serving). Drafts in "drafts in progress" never serve.
  • Billing summary: available balance vs daily budget → days of runway.

Post-launch (after 48 h)

  • Conversion actions left "Pending"? → the AW- tag on the site is not this account's, or the trigger page is never reached. Page-load actions are the quickest proof the tag works.
  • Clicks > 0 but GA4 shows no google / cpc sessions? → gclid is being stripped (page_location), or auto-tagging is off in Ads (Admin → Account settings → Auto-tagging).
  • Imported purchases in GA4 but not in Ads → import action not Primary, or the purchase happened outside the click-through window.

Report format

Lead with what was broken and is now fixed (with commit/links), then status per layer as a table, then "only you can do" items (top up, identity verification, publish, deletions), then "check again on <date>" items.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free