Skip to main content
skill Claude Code CX Codex Cursor GitHub Copilot

fixing-metadata — Claude Code skill: audit and fix HTML metadata

Audits page metadata and fixes what's broken: titles, descriptions, canonical URLs, Open Graph, Twitter cards, icons and JSON-LD.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add ibelick/ui-skills --skill fixing-metadata -a claude-code
Open on GitHub

What fixing-metadata does

The skill walks your pages and looks for missing or wrong metadata: title, description, canonical, OG tags. It then checks them against priority rules, starting with duplicates and indexing problems before anything cosmetic. It also makes sure title, description, canonical and og:url agree with each other.

Inside is a five-step workflow plus a rule reference split into eight categories: correctness and duplication, title and description, canonical and indexing, social cards, icons and manifest, structured data, locale and alternates, tool boundaries. Social images have to use absolute URLs, and twitter:card is set to summary_large_image. JSON-LD gets added only when it maps to what the page actually renders.

Diffs stay small. The skill won't refactor unrelated code or swap your framework, and it follows whatever metadata pattern the project already uses — the Next.js metadata API, react-helmet, or a hand-written <head>. Social cards are meant to be checked on a real URL, so localhost won't tell you much.

What it helps with

  • find pages with no title or meta description
  • strip duplicate canonical and robots tags
  • fix broken link previews on social platforms
  • set absolute URLs for og:image
  • match og:url against the canonical URL

Example prompts for the agent

  • audit the metadata on the blog pages and fix duplicate canonicals
  • the homepage social card shows a broken image — sort out og:image
  • add proper Open Graph and Twitter tags for the new landing page

How to install fixing-metadata

1Skills CLI (Vercel)

npx skills add ibelick/ui-skills --skill fixing-metadata -a claude-code

2Manually: put the files in place

~/.claude/skills/fixing-metadata/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/fixing-metadata/

From the same set5

The whole set →

Similar

brand-voice affaan-m Extracts an author's writing style from real publications and turns it into a reusable profile for drafts and outreach. Useful when you need consistent tone without typical AI cliches. Claude Code CX Codex Cursor +1 4,900installs content-engine affaan-m This skill turns articles, notes, and demos into content for X, LinkedIn, TikTok, YouTube, and newsletters. Use it for posts, scripts, or content calendars. Claude Code CX Codex Cursor +1 4,700installs marketing-campaign affaan-m Plans and runs multi-channel marketing launches. It handles audience research, campaign angles, landing page copy, email sequences, and social posts. Claude Code CX Codex Cursor +1 4,600installs content-creation official anthropics This skill helps draft marketing content for blogs, social posts, newsletters, landing pages, and more. Use it when channel formatting, SEO copy, headlines, or CTAs matter. Claude Code CX Codex Cursor +1 4,900installs keyword-research every-app Finds search query opportunities via OpenSEO MCP tools, evaluates SERP metrics, and saves vetted keywords to the project. Claude Code CX Codex Cursor +1 6,700installs competitor-analysis every-app A skill for analyzing one competitor's organic footprint: keywords, content themes, backlinks, and gaps. Delivers a report with a winnable action plan. Claude Code CX Codex Cursor +1 6,600installs seo-audit coreyhaines31 Diagnoses technical SEO issues, indexing drops, and metadata errors. Identifies why your site is losing organic search rankings in Google. Claude Code CX Codex Cursor +1 236,800installs brand claudekit Keeps brand voice, visual rules and design tokens in one markdown file, then feeds them to your agent's prompts and to your CSS variables. Claude Code CX Codex Cursor +1 39,500installs

Questions about fixing-metadata

How do I install fixing-metadata in Claude?

Run npx skills add ibelick/ui-skills --skill fixing-metadata -a claude-code. The skill lands at ~/.claude/skills/fixing-metadata/SKILL.md.

What does fixing-metadata do?

It audits HTML metadata and fixes what it finds: titles, descriptions, canonical URLs, Open Graph, Twitter cards, favicons, JSON-LD and robots directives.

How is fixing-metadata different from an SEO plugin?

It doesn't migrate frameworks or touch unrelated code, and it follows the metadata pattern already in your project. Changes stay scoped to metadata.

Source text

SKILL.md · MIT
---
name: fixing-metadata
description: >
  Audit and fix HTML metadata including page titles, meta descriptions, canonical URLs, Open Graph
  tags, Twitter cards, favicons, JSON-LD structured data, and robots directives. Use when adding
  SEO metadata, fixing social share previews, reviewing Open Graph tags, setting up canonical URLs,
  or shipping new pages that need correct meta tags.
version: 1.0.1
license: MIT
---

## Workflow

1. Identify pages with missing or incorrect metadata (titles, descriptions, canonical, OG tags)
2. Audit against the priority rules below — fix critical issues (duplicates, indexing) first
3. Ensure title, description, canonical, and og:url all agree with each other
4. Verify social cards render correctly on a real URL, not localhost
5. Keep diffs minimal and scoped to metadata only — do not refactor unrelated code
## when to apply

Reference these guidelines when:
- adding or changing page titles, descriptions, canonical, robots
- implementing Open Graph or Twitter card metadata
- setting favicons, app icons, manifest, theme-color
- building shared SEO components or layout metadata defaults
- adding structured data (JSON-LD)
- changing locale, alternate languages, or canonical routing
- shipping new pages, marketing pages, or shareable links

## rule categories by priority

| priority | category | impact |
|----------|----------|--------|
| 1 | correctness and duplication | critical |
| 2 | title and description | high |
| 3 | canonical and indexing | high |
| 4 | social cards | high |
| 5 | icons and manifest | medium |
| 6 | structured data | medium |
| 7 | locale and alternates | low-medium |
| 8 | tool boundaries | critical |

## quick reference

### 1. correctness and duplication (critical)

- define metadata in one place per page, avoid competing systems
- do not emit duplicate title, description, canonical, or robots tags
- metadata must be deterministic, no random or unstable values
- escape and sanitize any user-generated or dynamic strings
- every page must have safe defaults for title and description

### 2. title and description (high)

- every page must have a title
- use a consistent title format across the site
- keep titles short and readable, avoid stuffing
- shareable or searchable pages should have a meta description
- descriptions must be plain text, no markdown or quote spam

### 3. canonical and indexing (high)

- canonical must point to the preferred URL for the page
- use noindex only for private, duplicate, or non-public pages
- robots meta must match actual access intent
- previews or staging pages should be noindex by default when possible
- paginated pages must have correct canonical behavior

### 4. social cards (high)

- shareable pages must set Open Graph title, description, and image
- Open Graph and Twitter images must use absolute URLs
- prefer correct image dimensions and stable aspect ratios
- og:url must match the canonical URL
- use a sensible og:type, usually website or article
- set twitter:card appropriately, summary_large_image by default

### 5. icons and manifest (medium)

- include at least one favicon that works across browsers
- include apple-touch-icon when relevant
- manifest must be valid and referenced when used
- set theme-color intentionally to avoid mismatched UI chrome
- icon paths should be stable and cacheable

### 6. structured data (medium)

- do not add JSON-LD unless it clearly maps to real page content
- JSON-LD must be valid and reflect what is actually rendered
- do not invent ratings, reviews, prices, or organization details
- prefer one structured data block per page unless required

### 7. locale and alternates (low-medium)

- set the html lang attribute correctly
- set og:locale when localization exists
- add hreflang alternates only when pages truly exist
- localized pages must canonicalize correctly per locale

### 8. tool boundaries (critical)

- prefer minimal changes, do not refactor unrelated code
- do not migrate frameworks or SEO libraries unless requested
- follow the project's existing metadata pattern (Next.js metadata API, react-helmet, manual head, etc.)

## review guidance

- fix critical issues first (duplicates, canonical, indexing)
- ensure title, description, canonical, and og:url agree
- verify social cards on a real URL, not localhost
- prefer stable, boring metadata over clever or dynamic
- keep diffs minimal and scoped to metadata only

© ibelick · MIT · ibelick/ui-skills

See also

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor