Fix

Skill Sandy-zippy/claude-code-seo-stack/skills/fix

Execute the diagnosis - write the JSON-LD schema, the llms.txt, the copy changes and the answer content, then push live via a CMS MCP if one is connected (WordPress/Shopify/Webflow/Sanity). Use when asked to "fix it", "run the fix skill", or implement the diagnosis report.From its SKILL.md

Install

npx -y skills add Sandy-zippy/claude-code-seo-stack --skill fix

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.6 KB, 557 tokens by cl100k_base, as published. Nobody here has run it

Fix

Stop at nothing short of done. This skill writes every artifact the diagnosis calls for and, if a CMS MCP is connected, publishes it.

Inputs

seo-reports/diagnosis-<date>.md (run diagnose first if missing).
Optional: a CMS MCP (WordPress/Shopify/Webflow/Sanity). Without one, output files + paste-ready instructions instead.

Steps

Schema: for each schema fix, write complete JSON-LD. Minimums:
- Home: Organization with name, url, logo, sameAs (LinkedIn, Instagram, Google Business, directories — corroboration is the point)
- Service/product pages: Service/Product with provider, area served, offers
- Q&A-style content: FAQPage mapped to the real on-page questions Validate structure mentally against schema.org; no invented properties.
llms.txt: write one at the root. Plain markdown: who the brand is, what it does, proof (real numbers only — pull from the site, never invent), key pages with one-line descriptions, how to cite. Short beats long.
Copy changes: for each copy fix, produce a before/after diff of the exact sentences. Rules: claims must be verifiable, answer-shaped (front-load the answer, then support), no superlatives without proof.
Answer content: for each unanswered AI question in the diagnosis, draft the answer section (150-300 words): question as H2, direct answer in the first sentence, then evidence. Mark which page each section belongs on.
Publish (only if a CMS MCP is connected): push schema and copy changes page by page. After each push, fetch the live page and verify the change actually rendered. Anything that fails verification gets reported, not assumed.
Change log: append every change (or every prepared-but-not-pushed file) to seo-reports/changes-<date>.md so the weekly watchdog can correlate changes with movement.

Output

FIX RUN: <domain> — <date>
PUSHED LIVE: <n> changes (verified) / PREPARED: <n> files awaiting manual upload

1. <change> — <pushed+verified | prepared at path>
...
NEXT: set the weekly watchdog (templates/weekly-watchdog.md) to track movement.

Rules

Never push without verifying the render afterwards.
Never write a claim into copy or llms.txt that the site/user cannot back up.
Respect the diagnosis ranking — implement in order, top impact first.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.