Guide · Agent skill

How to use the Google Consent Mode v2 Auditor agent skill

Updated 2026-10-29 · written with AI from the project's own documentation and checked against our catalogue data

The consent-mode skill is an agent skill that gives an AI agent a checklist-style guide for Google Consent Mode v2, a Google system that changes how analytics and advertising tags behave depending on whether a visitor agreed to cookies. It is aimed at web developers, marketers and site owners who want to check their consent tags or look at whether they slow pages down. After reading this guide you will know what the skill does, how to install it, how to ask for it in plain language and what to expect from it.

Google Consent Mode v2 Auditorthedaviddias/Front-End-Checklist · MIT Download the skill

Quick start

1Download the ZIP. Download the consent-mode skill ZIP from the page where you found this guide.
2Unzip into a skills folder. Unzip it into ~/.claude/skills/ for personal use, or into your project's .claude/skills/ folder for one project.
3Or upload in the Claude apps. In the Claude apps, go to Settings → Capabilities → Skills and upload the ZIP.
4Describe the task. Ask in plain language, naming consent mode and the problem, such as slow loads or wrong consent states.
5Ask for evidence. Request the measurement method used (DevTools, Lighthouse or field data) and the exact file or request involved.
6Review before changing. Read the findings, test any proposed change, and keep legal decisions with a qualified lawyer.

What is an agent skill, and what does this one do?

An agent skill is a folder of instructions that an AI agent loads when a task matches the skill's description. This skill is defined in a SKILL.md file and points to a reference file, references/rule.md, for implementation details and code examples.

Once installed, the agent is guided to check that Google Consent Mode v2 is set up correctly and sends the right consent states. When you update Google Tag Manager or gtag.js, it points to the ad_user_data and ad_personalization consent types. Its trigger description focuses on auditing slow page loads, heavy assets and rendering delays related to consent mode, and it asks the agent to confirm the real bottleneck in DevTools, Lighthouse or field data before recommending changes.

Checks that consent states are sent correctly
Points to ad_user_data and ad_personalization when updating GTM or gtag.js
Looks for slow loads, heavy assets and rendering delays linked to consent tags
Asks for a named measurement method before recommending a fix

Download the skill ZIP from the page where you found this guide. Then choose one of the two install routes below. The source does not give a command-line install method, so the steps below use only the folders and menu names listed by the source.

For a personal install that works across your projects, unzip the download into ~/.claude/skills/. To limit the skill to one project, unzip it into that project's .claude/skills/ folder instead.

In the Claude apps you can also upload the ZIP under Settings → Capabilities → Skills.

Personal: unzip into ~/.claude/skills/
Single project: unzip into your project's .claude/skills/ folder
Claude apps: Settings → Capabilities → Skills, then upload the ZIP

Check what the skill needs before you start

The skill lists no required tools or packages. It does not run scripts and does not use the network, according to its metadata. It is a set of instructions, so the agent reads your code and any data you provide, and the audit itself depends on the tools you use (DevTools, Lighthouse or field data).

Requires: nothing listed
Runs scripts: no
Uses network: no
Licence: MIT

Trigger the skill with natural requests

You do not need a special command. Describe the task in plain language and mention consent mode or the symptom you are seeing. The agent should match the request to the skill's description.

Good requests name the problem and the evidence you want. For example, you might ask the agent to check whether a page with a consent banner is loading slowly because of consent tags, or to confirm that Consent Mode v2 sends the right states after a user accepts cookies.

Check whether our Google Consent Mode v2 setup sends the correct consent states.
Our checkout page loads slowly after adding the consent banner. Is the consent tag the cause? Confirm it before suggesting changes.
Update our GTM container so ad_user_data and ad_personalization follow the visitor's consent choice.
Review the loading behaviour of our consent setup and tell me which files or requests add unnecessary cost.

The reference file shows the default consent setup. It should sit as high as possible in the head of the page, before any GTM or gtag.js scripts. The example below sets all types to denied and includes a wait_for_update value, as copied from the source.

When a user accepts cookies through your consent management platform (CMP), the source describes updating consent after that user action. Ask the agent to check both the default state and the update.

<script>
  window.dataLayer = window.dataLayer || [];
  function gtag(){dataLayer.push(arguments);}

  // Default all to denied
  gtag('consent', 'default', {
    'ad_storage': 'denied',
    'ad_user_data': 'denied',
    'ad_personalization': 'denied',
    'analytics_storage': 'denied',
    'wait_for_update': 500
  });
</script>

Read the results and verify before changing code

The skill asks the agent to verify the actual bottleneck before recommending changes. Expect the agent to name the files, requests or rendering steps it reviewed and the method it used to confirm an issue.

Keep the final decision with your team. Ask the agent to show the evidence from DevTools, Lighthouse or field data, and check any proposed change in a test environment first.

Ask which measurement method was used
Ask for the exact file, request or rendering step involved
Test proposed changes before deploying them

Safety notes and limits

The skill is a technical guide. It does not give legal advice, and its wording on GDPR and other regulations is general. Talk to a lawyer before making compliance decisions.

The skill does not apply to sites with no Google tags or no consent banner, because its guidance is about Consent Mode v2 specifically. It runs no scripts and makes no network calls itself, so any checks happen through the tools you or the agent use.

Not legal advice on GDPR or other laws
Does not apply to sites without Google tags or a consent banner
No scripts run and no network use, per its metadata

What people use it for

Checking that a site's Consent Mode v2 tags send the correct consent states
Updating a GTM container so ad_user_data and ad_personalization respond to the visitor's choice
Investigating whether a consent banner or consent tags slow down page loads
Reviewing which assets or requests add unnecessary network or rendering cost around consent tags

Frequently asked questions

How do I install the consent-mode agent skill?

Download the ZIP from the page where you found this guide and unzip it into ~/.claude/skills/ for personal use, or into a project's .claude/skills/ folder. You can also upload the ZIP in the Claude apps under Settings → Capabilities → Skills.

Does the consent-mode skill require anything else to run?

The skill lists no requirements. Its metadata says it runs no scripts and uses no network. Any measurement, such as DevTools or Lighthouse, is done with the tools you choose.

Is the consent-mode skill free to use?

The skill is published under the MIT licence, as listed on its page. Check the licence file in the download for the full terms.

Does this skill make my site GDPR compliant?

No. The skill is a technical guide to Google Consent Mode v2 and its wording is general. It does not replace advice from a lawyer on GDPR or other regulations.

Will the skill work if my site has no consent banner or Google tags?

Not really. The guidance applies to sites that use Google Consent Mode v2 with a consent banner, so it is not useful for sites without Google tags or a consent banner.

How does this skill compare with a general Lighthouse audit?

A general Lighthouse audit reports on many page aspects. This skill is narrower: it focuses on Consent Mode v2 setup and on whether consent-related tags cause slow loads or heavy assets, and it asks the agent to confirm any bottleneck with DevTools, Lighthouse or field data.

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 or site owner who needs to check Google Consent Mode v2 tags or find out whether consent tags slow your pages. Pick a different tool, or get legal advice, if you need a compliance decision or a general site audit that covers far more than consent tags.

Download the skill