AugmentClaude

Analytics Audit

Review SDK configuration, event naming, funnel tracking, and dashboards before launch.

Installation

  1. Make sure Claude is on your device and in your terminal.

    Skills load from ~/.claude/skills/ when Claude Code starts up β€” so you need it on your machine first. If you don't have it yet, install it once with the command below, then run claude in any terminal to verify.

    One-time setup
    npm i -g @anthropic-ai/claude-code

    Already have it? Skip ahead.

  2. Paste into Claude Code or into your terminal.

    This copies the whole skill folder into ~/.claude/skills/analytics-audit-me2resh/ β€” the SKILL.md plus any scripts, reference docs, or templates the skill ships with. Works for every skill.

    What this command does
    1. Makes a new, empty temporary folder.
    2. Downloads the skill's folder, .claude/skills/analytics-audit, from github.com/me2resh/apexyard (branch main), the GitHub repository it's listed from. You get the version that is there now, not a copy we keep.
    3. Copies that folder to ~/.claude/skills/analytics-audit-me2resh/.
    4. Deletes the temporary folder.

    Installing runs none of the skill's files; it only downloads and copies them. When you later use the skill, Claude follows its SKILL.md and may run scripts that come with it. Read them first.

    How we check skills, and what we don't Β· Report this skill

    Faster alternative (instruction-only skills)

    Skips the clone and grabs only the SKILL.md file. Don't use this if the skill ships Python scripts, reference markdowns, or asset templates β€” they won't be downloaded and the skill will fail when it tries to load them.

    Quick install (SKILL.md only)
    Sign up to copy
  3. Restart Claude Code.

    Quit and reopen Claude Code (or any other agent that loads from ~/.claude/skills/). New skills are picked up on startup.

  4. Just ask Claude.

    Skills auto-activate when your request matches the skill's description β€” no slash command needed. Trigger phrases live in the skill's own frontmatter; you can read them in the β€œWhat this skill does” section above.

Prefer to read the source first? Open on GitHub.

When Claude uses it

Analytics audit β€” SDK config, event naming, funnel completeness, dashboards. Deep-dive for /launch-check analytics.

What this skill does

Writing: read docs/skills/shared/writing-rule.md before you write a durable artifact.

/analytics-audit β€” Event Taxonomy & Coverage

Deep-dive analytics analysis. Checks that tracking is configured, events follow a naming convention, key user funnels are instrumented, and dashboards exist. Invoke when /launch-check's analytics row shows WARN or FAIL.

Process

Step 1: SDK detection

Grep for analytics SDK initialization:

  • Google Analytics / GA4: gtag, G-, UA-, analytics.js, @google-analytics
  • Mixpanel: mixpanel.init, @mixpanel
  • Amplitude: amplitude.init, @amplitude
  • PostHog: posthog.init, posthog-js
  • Plausible: plausible.io, data-domain
  • Segment: analytics.load, @segment
  • Custom: any track(, capture(, logEvent( patterns

Step 2: Event inventory

Find all tracking calls in the codebase and list them:

  • Event name
  • Where it fires (file + component/handler)
  • Properties sent with the event
  • Whether it has a consistent naming convention (snake_case, camelCase, verb_noun)

Step 3: Funnel coverage

Identify the core user funnels and check if each step is tracked:

FunnelSteps to check
Signuppage_view β†’ form_start β†’ form_submit β†’ signup_complete
Activationfirst_login β†’ key_action_completed β†’ aha_moment
Purchase (if applicable)pricing_view β†’ plan_select β†’ checkout_start β†’ payment_complete
Retentionreturn_visit β†’ feature_usage β†’ session_duration

Step 4: Output

ANALYTICS AUDIT β€” <project> @ <sha>

SDK: <GA4 / Mixpanel / PostHog / none>
Total events found: <N>
Naming convention: <consistent / inconsistent / none>

| # | Area | Status | Finding |
|----|------|--------|---------|
| E1 | SDK | PASS | GA4 initialized in _app.tsx |
| E2 | Events | WARN | 8 events found, 3 use camelCase, 5 use snake_case |
| E3 | Signup funnel | FAIL | form_submit tracked but signup_complete missing |
| E4 | Dashboard | WARN | No dashboard URL found in config or docs |

Event inventory:
  signup_start (pages/signup.tsx:42)
  form_submit (components/SignupForm.tsx:89)
  page_view (auto, GA4 enhanced measurement)
  ...

Persist the run + render trend

After printing the findings table, persist via the shared audit-history lib so the analytics trend across runs becomes legible. See docs/agdr/AgDR-0019-audit-artefact-persistence.md.

Resolve project name + score + verdict

<project-name> from apexyard.projects.yaml (or basename + /handover reminder if unregistered).

Score: score = max(0, 100 - 25*critical - 10*high - 3*medium - 1*low). Verdict by worst-severity: critical/high β†’ fail, medium β†’ conditional, low/none β†’ pass.

Persist + render

source "$(git rev-parse --show-toplevel)/.claude/hooks/_lib-audit-history.sh"

# Lowercase severity in the payload β€” the lib expects critical/high/medium/low/info.
payload=$(mktemp); cat > "$payload" <<'EOF'
{
  "schema_version": 1,
  "findings": [
    {"id": "E2", "severity": "medium", "status": "open", "summary": "Mixed event-naming conventions (snake_case + camelCase + Title Case)"},
    {"id": "E3", "severity": "high",   "status": "open", "summary": "signup_complete event missing β€” can't measure conversion"},
    {"id": "E4", "severity": "medium", "status": "open", "summary": "No dashboard URL found in config or docs"}
  ]
}
EOF

# Body: per templates/audits/analytics-audit.md
body=$(mktemp); cat > "$body" <<'EOF'
... (filled-in body β€” findings table + Event taxonomy gap + Recommended priority) ...
EOF

ts=$(date -u +"%Y-%m-%dT%H:%M:%SZ")
audit_run_persist "<project-name>" "analytics-audit" "$ts" "fail" 60 "$body" < "$payload"
rm -f "$payload" "$body"

audit_render_trend "<project-name>" "analytics-audit" 5

Opt-in commit

touch projects/<name>/audits/analytics-audit/.audit-history-tracked

Rules

  1. Auto-PASS for non-user-facing projects. CLIs, libraries, and internal tools don't need analytics.
  2. Don't prescribe a specific SDK. Note what's configured, check coverage, suggest improvements.
  3. Flag inconsistent naming β€” mixed conventions make dashboard queries painful.
  4. Privacy-aware. Flag if PII (email, name, IP) is being sent in event properties.
  5. Always persist via the lib. The persist step runs regardless of opt-in commit state.
  6. Severity vocabulary in the JSON is lowercase. The lib expects critical/high/medium/low/info.

Part of ApexYard β€” multi-project SDLC framework for Claude Code Β· MIT.

Related skills