Guide · Agent skill
How to use the Descriptive Anchor Text Audit agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The anchor-text skill is a written checklist that teaches an AI agent to review the links on a website and flag link text that does not say where the link goes, such as 'click here' or 'read more'. This guide explains what the skill makes the agent do, how to install it, how to ask for an audit in plain language, and what to watch for. It is aimed at readers who have never used agent skills before.
Descriptive Anchor Text Auditthedaviddias/Front-End-Checklist · MIT Download the skill Quick start
- 1Download the ZIP. Get the anchor-text skill as a ZIP from its page in the directory.
- 2Unzip into a skills folder. Extract it to ~/.claude/skills/ for personal use or to .claude/skills/ inside a project.
- 3Or upload in the Claude apps. In the Claude apps, go to Settings → Capabilities → Skills and upload the ZIP.
- 4Ask for an audit in plain language. Describe the SEO or accessibility review you want and name the pages or templates to check.
- 5Review the suggested wording. Check each proposed anchor text against the target page before you change anything.
What the anchor-text skill makes the agent do
Descriptive anchor text is the clickable text of a link. The skill tells the agent that this text should clearly indicate what a reader will find when they follow the link. Search engines use that context to understand the linked page, and screen reader users often navigate by listing all the links on a page, so vague wording makes that list hard to use.
Once the skill is installed and you ask for an SEO or accessibility audit, the agent looks for generic link phrases, suggests clearer wording that includes relevant terms for the destination page, and checks the rendered page output rather than only the source files.
- Flags generic phrases such as 'click here' or 'read more'
- Suggests replacement anchor text that describes the target page
- Warns against over-optimising internal links with keyword stuffing
- Asks the agent to verify the rendered HTML and HTTP response, not just source files
Install the anchor-text skill
Download the skill as a ZIP from the skill's page in the directory. Then choose one of two routes.
For Claude Code, unzip the file into your personal skills folder at ~/.claude/skills/, or into a project's .claude/skills/ folder if you want the skill to apply only to that project. For the Claude apps, upload the ZIP under Settings → Capabilities → Skills.
- Personal install: ~/.claude/skills/
- Project install: .claude/skills/ inside your project
- Claude apps: Settings → Capabilities → Skills, then upload the ZIP
Trigger the skill with natural requests
You do not need a special command. The skill's description is about auditing SEO, crawlability, structured data and indexability, so asking for those things in ordinary language is usually enough. Name the pages, templates or site you want reviewed so the agent knows where to look.
Good requests state the goal and the scope. For example: review the navigation and footer templates for vague link text, or check the blog post links for wording that does not describe the target.
- Audit the header and footer templates for generic link text like 'click here'.
- Check our internal links on the services pages for descriptive anchor text.
- Review this HTML file for links that do not say where they go, and suggest better wording.
- Look at the rendered output of the product category page and flag unclear links for accessibility.
What you get back and what it needs
The skill is a single written guide. It gives the agent a checklist (Quick Reference, Check, Fix, Explain and Code Review) and a reference file with a good and bad HTML example and the reasons the rule matters. The agent applies that guidance to whatever pages or code you provide. The source does not describe any executable tools, so the agent relies on tools it already has to read files or pages.
The skill lists no requirements, runs no scripts and does not use the network.
- Requirements: none listed
- Runs scripts: no
- Uses network: no
- Files included: SKILL.md and references/rule.md
Safety notes and limits
Because the skill runs no scripts and uses no network access, it does not itself fetch or change anything on your site. Any reading of pages or editing of files comes from the agent's own tools and your permissions.
The skill is not a crawler. It will not scan every page of a large site automatically. If you need that, the agent would need its own tooling for crawling. The skill is also a best-practice checklist, not an accessibility or legal compliance standard, so it does not certify that a site meets any rule.
- Not a crawler: large-scale scanning needs separate tools
- Not a compliance certificate: treat results as best-practice guidance
- Review suggested wording before publishing, since the skill gives guidance only
Tips for better results
Ask the agent to check the rendered HTML and the HTTP response where you can, because the skill itself stresses verifying the final page output rather than relying only on source files. Share the specific templates or routes that produce links, so suggestions can be tied to real code.
Keep anchor text natural. The skill warns against stuffing keywords into internal links, so aim for wording a reader would find clear on its own.
- Point the agent at templates or routes, not just the whole project
- Ask for before-and-after wording so you can review each change
- Use the skill for content editors too, when you want clearer link wording
What people use it for
- Reviewing site navigation and footer templates for 'click here' or 'read more' links
- Checking internal links on service or product pages for descriptive wording
- Improving link text in blog posts and help articles written by content editors
- Preparing accessibility notes that flag links which do not make sense out of context
Frequently asked questions
Is the anchor-text skill free?
The skill is published under the MIT licence, as stated on its page. Check the licence file in the repository for the exact terms before reusing it.
What do I need to use the anchor-text skill?
The skill lists no requirements. You install it by unzipping it into a skills folder or uploading it in the Claude apps, then ask the agent for an audit.
Does the skill run scripts or access the internet?
No. The skill runs no scripts and does not use the network. It is a written guide that the agent applies to files or pages you provide.
Can the skill crawl my whole website?
No. The skill includes no crawler, so it cannot scan every page automatically. For large-scale scanning you would need separate tools that the agent can use.
Where do I install the skill in Claude Code?
Unzip it into ~/.claude/skills/ for personal use, or into .claude/skills/ inside a specific project.
Is this skill an accessibility certification?
No. It is a best-practice checklist for descriptive link text, not an accessibility audit standard, so it does not certify legal compliance.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use this skill if you are a web developer, SEO specialist or content editor who wants an agent to catch vague link wording and suggest clearer alternatives. Pick something else if you need automated scanning of every page at scale or a formal accessibility or legal compliance report.
Download the skill