cookie-audit

cookie-audit is a skill for Claude Code from capazme/mcp-legal-it. It costs 258 tokens per session (2,529 once invoked), scanned A, original, Apache-2.0.

A website cookie audit tool that checks the cookies and tracking services a public site actually loads before and after consent.

In plain words
What is it for?
Use it to inventory cookies and trackers, check GDPR and Italian cookie-law compliance, and prepare remediation materials or a report.
Why use it?
It reveals differences between a site's stated cookie policy and its real behaviour, including possible tracking before consent.

Skill for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT} variable. Also seen: reads .claude/ paths.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the legal-it plugin — 23 skills, 8 commands, 6 agents, 1 hook, 1 MCP server shipped together

Good fit Use it to inventory cookies and trackers, check GDPR and Italian cookie-law compliance, and prepare remediation materials or a report.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add capazme/mcp-legal-it
Claude Code
/plugin install legal-it

Made for: Claude Code.

Or install legal-it, the plugin that ships this one along with the rest of its 23 skills, 8 commands, 6 agents, 1 hook, 1 MCP server.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for cookie-audit

README.md
[![agentmods](https://agentmods.dev/badge/skills/capazme/mcp-legal-it/cookie-audit/github.svg)](https://agentmods.dev/skills/capazme/mcp-legal-it/cookie-audit)
Your own site
<a href="https://agentmods.dev/skills/capazme/mcp-legal-it/cookie-audit"><img src="https://agentmods.dev/badge/skills/capazme/mcp-legal-it/cookie-audit/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for cookie-audit

Your own site · 80×15
<a href="https://agentmods.dev/skills/capazme/mcp-legal-it/cookie-audit"><img src="https://agentmods.dev/badge/skills/capazme/mcp-legal-it/cookie-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 258 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,529 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00258 $0.02529
Opus 5 $0.00129 $0.01264
Sonnet 5 $0.00052 $0.00506
Haiku 4.5 $0.00026 $0.00253

Measured 11d ago against content hash bd489e5a1877, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

cookie-audit scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.

The scan reads SKILL.md. This mod also ships 2 executable files (scripts/generate_report.js, scripts/inspect_container.sh), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

Se il sito usa GTM (container `GTM-XXXXXXX`), **scarica il container lato server** con `curl` (nessun ad-blocker) ed estrai gli ID reali dei tag. È la fonte più affidabile per sapere *cosa* il sito installa davvero:
plugin/skills/cookie-audit/SKILL.md · 92 lines

How it starts

The opening of the file, as written. The whole thing — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Produce un'analisi tecnico-legale dei cookie e degli strumenti di tracciamento di un sito pubblico, con la stessa rigorosità di un accertamento privacy: non ci si limita a leggere la cookie policy dichiarata (spesso auto-generata e imprecisa), ma si osserva il comportamento reale del sito nel browser, distinguendo ciò che accade prima e dopo il consenso.

Cosa produce

  1. Tabella cookie unica — nome · fornitore/dominio · finalità · durata · categoria · stato (osservato / atteso / condizionale).
  2. Inventario tracker — CMP usata, container GTM, ID di misurazione/conversione reali (GA4, Ads, Meta, LinkedIn…).
  3. Valutazione di conformità — contro il Provvedimento del Garante 10.6.2021, il GDPR e l'art. 122 Cod. Privacy (vedi references/compliance-checklist.md).
  4. Report Word su richiesta (scripts/generate_report.js).
  5. Remediation su richiesta — bozza di cookie policy allineata allo stack reale + checklist di adeguamento.

Principio cardine

Il test decisivo del diritto italiano/UE è: nessun cookie non tecnico (né chiamata a terze parti non necessarie) prima di un consenso libero e specifico (art. 122 Cod. Privacy, che recepisce l'art. 5(3) Dir. ePrivacy). Tutto il metodo ruota attorno a verificare empiricamente questo.


Metodo

Setup del browser

Serve un MCP browser che sappia (a) navigare, (b) eseguire JavaScript nella pagina, (c) leggere le richieste di rete. Vanno bene chrome-devtools (mcp__chrome-devtools__*) o claude-in-chrome (mcp__claude-in-chrome__*); carica i tool via ToolSearch se sono deferred. Preferire chrome-devtools quando disponibile (contesto isolato + lettura rete più semplice).

Apri il sito in un contesto isolato/pulito (nessun cookie pregresso) — es. new_page({url, isolatedContext:"cookieaudit"}). Un contesto sporco falsa la Fase 1.

⚠️ Caveat ad-blocker (leggi sempre). Molti browser hanno uBlock Origin o simili. Un ad-blocker: blocca il CSS cookieblocker, nasconde cosmeticamente il banner (display:none), sostituisce gtm.js/gtag/js con uno stub e blocca gli endpoint analytics. Sintomi: _ga non compare dopo il consenso, gtm.js pesa pochi KB, richieste net::ERR_BLOCKED_BY_CLIENT. In tal caso le osservazioni pre-consenso restano valide, ma per l'inventario reale non fidarti del browser: usa la Fase 3-bis (container GTM lato server). Dichiara sempre questo caveat nel report e marca le voci non osservate come «Atteso».

Read the full file on GitHub · 92 lines

Files

What ships with it

5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 11d ago First seen · 92 lines · 258 tokens per session scan A bd489e5a1877

Subscribe to this mod's changes

cookie-audit is a skill published in the GitHub repository capazme/mcp-legal-it (15 stars, last pushed today), licensed Apache-2.0. It adds 258 tokens to every session and 2,529 once invoked, about $0.0013 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.