Skip to main content
skill CX Codex

title-tag — Claude Code skill: what it does, how to install

Audits and writes HTML title tags with checks for search intent, H1 alignment, and keyword cannibalization.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add kostja94/marketing-skills --skill title-tag -a claude-code
npx skills add kostja94/marketing-skills --skill title -a claude-code
Open on GitHub

What title-tag does

The skill handles drafting and reviewing the HTML <title> tag. It reads contextus.md in the project root when present, checking page role, primary search intent, current H1, and template brand suffixes. If a title fits a sibling page without edits, the Swap Test requires adding page-specific facts before finalizing the copy.

It treats roughly 60 Latin characters for the title body as a soft warning rather than a rigid limit. Brand suffixes like | Brand stay separate from the main character target. When multiple URLs target the same search intent, the skill flags cannibalization and reports overlapping URLs, but it leaves site architecture and keyword assignments untouched.

What it helps with

  • draft page-specific HTML title tags
  • audit title alignment against the primary H1
  • detect search intent cannibalization across sibling URLs
  • separate brand suffixes from core title copy

Example prompts for the agent

  • Review the title tag for this product page against its H1
  • Write an SEO title tag keeping the brand suffix separate
  • Run the swap test on these sibling page titles

How to install title-tag

1Skills CLI (Vercel)

npx skills add kostja94/marketing-skills --skill title-tag -a claude-code
npx skills add kostja94/marketing-skills --skill title -a claude-code

2Manually: put the files in place

~/.claude/skills/title/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/title/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about title-tag

How to install title-tag in Claude Code?

Run npx skills add kostja94/marketing-skills --skill title-tag -a claude-code in your terminal.

What does the title-tag skill do?

It audits and writes HTML title tags, balances length warnings, verifies alignment with H1, and flags search cannibalization.

Does title-tag edit page headings directly?

No, it checks semantic alignment with the H1, but routing actual heading edits belongs to the heading-structure skill.

Source text

SKILL.md · MIT
---
name: title-tag
description: When the user wants to write, audit, or optimize an HTML title tag, SEO title, page title, SERP title, browser-tab title, duplicate title, or title rewrite issue. For a sitewide metadata lifecycle including implementation and production validation, use page-metadata.
metadata:
  version: 2.0.0
---

# SEO On-Page: Title Tag

Owns writing and review rules for the HTML `<title>` element. **page-metadata** owns discovery, implementation, and release verification.

## Inputs

Read root `contextus.md` when present. Confirm page type, unique page job, primary search intent, locale, current title and H1, verified differentiators, sibling-page titles, and whether a template automatically appends the brand.

## Rules

- Write a descriptive, concise, page-specific title that accurately represents the page.
- Put the primary search concept early when natural; do not keyword-stuff.
- Treat about 60 Latin characters for the title body as a default warning, not a hard limit. Pixel width, script, device, and query context affect display.
- A delimiter plus brand suffix such as `| Brand` is outside that title-body target. It may be truncated or rewritten; include it only when useful and never duplicate a template suffix.
- For CJK, RTL, or other scripts, review rendered width and natural language instead of applying a translated character formula.
- Title and H1 must serve the same page intent and promise, but need not use identical wording. Example: title `AI Image Generator`; H1 `Generate Images with AI`.
- Include only claims supported by the page. Do not add a year, number, superlative, or feature merely for click appeal.
- Apply the **Swap Test**: if the title fits a sibling page unchanged, add a verified page-specific fact or sharpen its task.
- Google may generate a different title link from the `<title>`, main visual title, headings, prominent text, anchors, `og:title`, or other signals. Optimize consistency of page identity, not control of the exact displayed string.

## Cannibalization

When multiple URLs target the same primary intent, report the affected URLs, overlap, and evidence. Do not merge pages, reassign keywords, change page roles, or alter information architecture.

## Output

- Recommended title, with the title body and optional brand suffix shown separately.
- A length or width warning when relevant, never a mechanical pass/fail.
- Brief rationale tied to intent and unique page facts.
- H1 semantic-alignment result; actual H1 edits route to **heading-structure**.
- Cannibalization warning when applicable.

## Primary reference

- [Google: influencing title links](https://developers.google.com/search/docs/appearance/title-link)

## Related Skills

- **page-metadata**, **meta-description**, **heading-structure**
- **localization-strategy**, **translation**, **google-search-console**

© kostja94 · MIT · kostja94/marketing-skills

Similar skills

Everything for SEO and GEO →

From the same set120

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