What are the critical metadata issues this skill prioritizes?

Critical issues include ensuring one metadata source per page, eliminating duplicate title/description/canonical/robots tags, using deterministic values, and sanitizing dynamic strings. The skill also prioritizes fixing incorrect indexing directives and ensuring canonical URLs point to the preferred page.

What should I verify before applying the skill's output?

You should verify generated code and dependencies before applying changes, and ensure environment-specific testing, security review, and user approval occur for any destructive or costly actions. Additionally, verify that social cards render correctly on a real URL rather than localhost.

When should I NOT use this skill?

Do not use this skill to refactor unrelated code or migrate metadata frameworks unless that was explicitly requested. Use it only when the task matches its scope and the local project context, and do not treat its guidance as a substitute for proper testing and security review.

Skill

Fix Website Metadata and SEO Issues

Name: Fixing Metadata
Availability: OnlineOnly
Author: Antigravity

This skill audits and fixes HTML metadata - titles, descriptions, canonical URLs, Open Graph, Twitter cards, favicons, and JSON-LD - by priority.

Get skill

Antigravity

Own this? Claim it

Spark score

out of 100

Updated 19 days ago

Version 14.1.0

Add to Favorites

Why it matters

Ensure your website's metadata is accurate, consistent, and optimized for search engines and social sharing. This skill identifies and corrects issues with titles, descriptions, canonical tags, and social card rendering.

Outcomes

What it gets done

Audit and fix missing or incorrect page titles and descriptions.

Ensure canonical URLs and indexing rules are correctly set.

Verify and correct Open Graph and Twitter card metadata for social sharing.

Standardize metadata across pages for consistency and searchability.

Install

Add it to your toolbox

Run in your project directory:

curl -fsSL https://spark.entire.vc/get/ag-fixing-metadata | bash

Overview

Fixing Metadata

This skill audits and fixes HTML metadata - titles, descriptions, canonical URLs, Open Graph/Twitter cards, favicons, and JSON-LD - organized into eight priority categories from critical duplication/indexing fixes down to locale metadata. Use it when adding, auditing, or fixing page metadata for SEO or social sharing. Keep changes scoped to metadata only, verify social cards on a real URL not localhost, and don't migrate the project's existing metadata framework unless asked.

What it does

This skill audits and fixes HTML metadata: page titles, meta descriptions, canonical URLs, Open Graph tags, Twitter cards, favicons/manifests, JSON-LD structured data, and robots directives. It organizes fixes by priority into eight rule categories: correctness and duplication (critical - one metadata source per page, no duplicate title/description/canonical/robots tags, deterministic values, sanitized dynamic strings), title and description (high - every page needs a title, consistent format, plain-text descriptions with no markdown), canonical and indexing (high - canonical points to the preferred URL, noindex only for private/duplicate/non-public pages, correct paginated-page canonical behavior), social cards (high - Open Graph title/description/image with absolute URLs, og:url matching canonical, appropriate og:type and twitter:card), icons and manifest (medium - cross-browser favicon, apple-touch-icon, valid manifest, intentional theme-color), structured data (medium - JSON-LD only when it clearly maps to real page content, never inventing ratings/reviews/prices), locale and alternates (low-medium - correct html lang, og:locale, hreflang only for pages that truly exist), and tool boundaries (critical - minimal, scoped changes; no framework or SEO-library migration unless requested; follow the project's existing metadata pattern, whether Next.js metadata API, react-helmet, or manual head tags).

The workflow: identify pages with missing or incorrect metadata, fix critical issues (duplicates, indexing) first, ensure title/description/canonical/og:url are internally consistent, verify social cards render correctly on a real URL rather than localhost, and keep diffs scoped to metadata only rather than refactoring unrelated code.

When to use - and when NOT to

Use this skill 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 metadata components, adding JSON-LD structured data, changing locale/alternate-language/canonical routing, or shipping new shareable/marketing pages. Per its own tool-boundaries guidance, use it only when the task matches its scope and the local project context, verify generated code and dependencies before applying changes, and do not treat its guidance as a substitute for environment-specific testing, security review, or user approval for destructive or costly actions - and don't use it as license to refactor unrelated code or migrate metadata frameworks unless that was actually requested.

Inputs and outputs

Input is a page or set of pages with metadata to audit (titles, descriptions, canonical/OG/Twitter tags, icons, structured data). Output is a metadata fix scoped to the identified issues, prioritized by the eight rule categories, with critical correctness/duplication and indexing issues addressed first.

Integrations

Applies to whatever metadata mechanism a project already uses - Next.js's metadata API, react-helmet, or manually authored <head> tags - without prescribing a specific framework migration.

Who it's for

Frontend and SEO engineers who need to audit or fix a page's HTML metadata (titles, social cards, structured data, canonical/indexing signals) correctly and with minimal, scoped changes.

FAQ

Common questions

Discussion