agentleFS
powerczukpowerczuk/seo-machine/skills/navboost-audit/SKILL.md

navboost-audit

Skill0 starsChanged 2 months agoMITView on GitHub

NavBoost engagement audit from GSC/GA4 — position-normalized CTR, pogo-stick proxy, engaged sessions, duration, branded and brand+topic search, Knowledge Panel — 70-point scorecard.

The file

Files
SKILL.md

/navboost-audit — Engagement signal audit

Target: $ARGUMENTS

Background, click typology, and baselines: ${CLAUDE_PLUGIN_ROOT}/skills/seo-fundamentals/references/navboost.md. Scorecard bands: ${CLAUDE_PLUGIN_ROOT}/skills/seo-fundamentals/references/scoring-spec.md.

Inputs

Domain; brand name(s) and variants; primary topic cluster(s). Data: Google Search Console and GA4 — via MCP if connected, otherwise ask for CSV exports or screenshots (GSC: queries and pages, 12 months; GA4: landing pages with sessions, engaged sessions, average engagement time). Modules without data are skipped and listed as "not run", never estimated.

Modules (10 points each)

  1. Position-normalized CTR — top 100 queries by impressions: actual CTR ÷ expected CTR for the position (baselines in navboost.md; adjust for SERP features). ≥1.2 STRONG · 0.8–1.2 AT BENCHMARK · <0.8 WEAK. Site score from the share of queries ≥1.0 (target ≥40% above benchmark). Flag queries where an AI Overview appeared before calling a CTR drop "weak copy".
  2. Pogo-stick proxy — top 20 organic landing pages: share of sessions <10 s with no interaction. <20% STRONG · 20–40% WATCH · >40% WEAK.
  3. Engaged sessions ratio — site-wide >60%, per page >50%.
  4. Session duration — top 20 landing pages, median engagement time; bounce-back-to-organic rate as a lastLongestClick failure proxy.
  5. Branded search volume — 12-month trend (GSC brand filter + Trends): growing / flat / declining.
  6. Brand + topic queries — GSC filter brand AND topic: the best available topical-authority proxy; growing / stable / declining.
  7. Knowledge Panel — present? complete (logo, description, sameAs-backed profiles, related topics)?

Output

  1. Scorecard: 7 modules × 10 = 70, with "not run" modules excluded from the total and the denominator shown.
  2. Verdict: STRONG 56–70 · GOOD 42–55 · WATCH 28–41 · WEAK 14–27 · CRITICAL 0–13 (pro-rate when modules are missing).
  3. Top-3 weak queries (meta rewrite candidates → /meta-tags).
  4. Top-3 weak landing pages (content-intent mismatch → /optimize).
  5. Branded and brand+topic trend summary (chart if data allows).
  6. Knowledge Panel gap report (→ Organization schema with sameAs, /schema-gen).
  7. Prioritized actions for the biggest NavBoost lift; offer a monthly re-run as a scheduled task.

Caveats: Chrome-side data is invisible to site owners; CTR baselines vary by vertical and SERP layout; AI Overviews divert clicks at stable positions — compare before/after the AI Overview appeared for the query.

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.