Skill · WordPress & Dev

Core Web Vitals Fixer for WordPress

Diagnoses the real cause of failing LCP, INP, and CLS on WordPress sites — and delivers prioritized, scored fixes with runnable code, including the non-obvious insight that consent-gated analytics loading is a measurable CWV fix, not just a privacy obligation.

Category: WordPress & Dev
Deliverable: 1 .skill bundle
Outputs: 6
Last updated: 15 Jun 2026

$11.99 One-time · lifetime updates

Get it on Agensi

Works in Claude Pro, Team, and Enterprise
Lifetime access to updates
Refundable for 30 days via the marketplace

Or get a free skill every month. Subscribers get one curated skill, free, every 1st. Pick yours →

StrategistKit Affiliate. Purchase happens on the marketplace, which handles payment, delivery and refunds.

Overview

What Core Web Vitals Fixer for WordPress does.

This skill traces failing LCP, INP, and CLS scores to their specific root causes in the WordPress/PHP/browser stack — not generic plugin recommendations. You supply a PageSpeed Insights URL, a raw PSI score, or a plain symptom description such as 'mobile LCP is 3.8 s after adding GA4'. The skill selects DIAGNOSE or FIX mode automatically, builds a ranked root-cause list scored on an anchored Impact-squared-over-Effort table, then delivers runnable before-and-after PHP and HTML patches for every finding: hero image preload hooks, font-display calibration, wp_get_attachment_image() with explicit dimensions, and a Consent Mode v2 deferred analytics pattern that eliminates bandwidth contention from an unconditionally loaded gtag.js.

One concrete example: a buyer pastes 'PSI mobile score dropped from 99 to 84, LCP went from 1.5 s to 3.8 s, TBT is still 0 ms.' The skill identifies the cause as gtag.js (approximately 146 KB) loading unconditionally in wp_head and competing with the hero image for the browser's narrow LCP bandwidth window — not a caching or server issue. Output includes an Impact²/Effort audit table ranking this finding first, a deferred Consent Mode v2 code snippet, a hero preload wp_head action, and a curl-based acceptance checklist so a second engineer can verify the result objectively without judgment calls.

Who it's for

WordPress developers and site owners who have PageSpeed Insights data showing red or amber CWV scores and need to know exactly which file, hook, or script is responsible — not a shortlist of plugins to try. Particularly useful for anyone who has already installed a caching plugin and still cannot clear LCP or CLS, or who noticed a score regression after adding analytics or a consent banner.

What you get

One skill. 6 outputs.

One .skill bundle. Run it on your material and it returns:

Two operating modes: DIAGNOSE (PSI/CrUX data or symptom → ranked root causes) and FIX (prioritized action plan with code).

Per-metric diagnosis trees for LCP, INP, and CLS — each with specific WordPress root causes and targeted fixes.

Runnable before→after code for every fix: hero image preload snippet, font-display: swap with fallback metric calibration, wp_get_attachment_image() with dimensions, Consent Mode v2 deferred analytics pattern.

Impact²/Effort scored audit table — so you fix the highest-ROI item first, not the most obvious one.

Verifiable acceptance checklist — every item has its own curl/PSI/DevTools verification method, so a second engineer can score the work objectively.

Block theme advantage section — explains the structural performance benefit of FSE themes and when migrating is worth considering.

How it works

Three steps. About two minutes.

Install

Add the .skill file to your Claude app. ~10 seconds.

Run it on your work

Invoke the skill and paste in your material.

Apply the output

Review, keep what works, and use it.

In depth

Why a Claude skill beats a prompt template.

A copy-paste prompt runs one static pass and stops. A skill is a bundled program — instructions, examples, and a workflow Claude runs as a unit: it asks for the right input, applies the same pattern every time, and returns the structured outputs above.

FAQ

Common questions.

What do I need to provide to get useful output?

A PageSpeed Insights URL, a copied PSI score breakdown, or a plain-language symptom description — for example, 'LCP is 4.2 s on mobile, CLS is 0.18, using Elementor on managed WordPress hosting with GA4.' The more detail you supply, the fewer clarifying questions the skill asks before delivering findings.

Does this skill generate real code or just advice?

It delivers runnable PHP and HTML for every fix: wp_head preload snippets, font-display swap with fallback metric calibration, wp_get_attachment_image() calls with explicit width and height, and a Consent Mode v2 deferred loading pattern. Every code block includes a before state and an after state so you can see exactly what changes.

Can it help if my site uses a block theme or full-site editing?

Yes. The skill includes a dedicated section explaining the structural performance advantages of block themes — per-page CSS loading, automatic font-display via theme.json — and assesses whether a theme migration would materially improve your scores given your current setup.

Does the skill cover all three Core Web Vitals metrics?

It covers LCP, INP, and CLS with separate diagnosis trees. Each tree maps symptoms to WordPress-specific root causes: for LCP, hero image preload and format issues; for INP, main-thread third-party scripts and heavy event handlers; for CLS, images without dimensions, late-injected banners, and font-swap shift.

What is the acceptance checklist and how do I use it?

At the end of FIX mode output, the skill produces a checklist where every item has its own verification method — a curl command, a PSI URL check, or a specific DevTools panel — so the results can be scored by a second engineer without relying on subjective assessment.

Skills used with this one.

WordPress & DevAgensi