lgpd-anonymization

lgpd-anonymization is a skill for Claude Code from goul4rt/lgpd-skills. It costs 90 tokens per session (767 once invoked), scanned A, original, MIT.

Guidance for removing or reducing the ability to identify people in data used for analytics or machine learning. Anonymization is intended to be irreversible by reasonable means, while pseudonymization replaces identity details but can be reversed with separate information.

In plain words
What is it for?
Use it to choose techniques such as tokenization, salted hashing, generalization, suppression, noise, k-anonymity, l-diversity, t-closeness, or differential privacy for analytics pipelines.
Why use it?
It helps reduce privacy risk and distinguish data that may fall outside LGPD from data that remains personal data.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the lgpd-skills plugin — 19 skills shipped together

Good fit Use it to choose techniques such as tokenization, salted hashing, generalization, suppression, noise, k-anonymity, l-diversity, t-closeness, or differential privacy for analytics pipelines.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/goul4rt/lgpd-skills/lgpd-anonymization
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add goul4rt/lgpd-skills --skill lgpd-anonymization
Clone the repo
git clone --depth 1 https://github.com/goul4rt/lgpd-skills

Made for: Claude Code.

Or install lgpd-skills, the plugin that ships this one along with the rest of its 19 skills.

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 lgpd-anonymization

README.md
[![agentmods](https://agentmods.dev/badge/skills/goul4rt/lgpd-skills/lgpd-anonymization/github.svg)](https://agentmods.dev/skills/goul4rt/lgpd-skills/lgpd-anonymization)
Your own site
<a href="https://agentmods.dev/skills/goul4rt/lgpd-skills/lgpd-anonymization"><img src="https://agentmods.dev/badge/skills/goul4rt/lgpd-skills/lgpd-anonymization/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 lgpd-anonymization

Your own site · 80×15
<a href="https://agentmods.dev/skills/goul4rt/lgpd-skills/lgpd-anonymization"><img src="https://agentmods.dev/badge/skills/goul4rt/lgpd-skills/lgpd-anonymization.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 90 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 767 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00090 $0.00767
Opus 5 $0.00045 $0.00383
Sonnet 5 $0.00018 $0.00153
Haiku 4.5 $0.00009 $0.00077

Measured 12d ago against content hash 92273bc421eb, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

lgpd-anonymization scanned grade A with 0 findings 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 12d ago.

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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/lgpd-anonymization/SKILL.md · 69 lines

How it starts

The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Anonimização e Pseudonimização

Diferença crítica

Anonimização (Art. 5º, XI) Pseudonimização (Art. 13, § 4º)
Reversível? Não, considerando meios técnicos razoáveis Sim, com info adicional segregada
Status LGPD Fora do escopo (Art. 12) Continua sendo dado pessoal
Uso típico Estatísticas públicas, datasets de pesquisa Analytics interna, logs, redução de exposição
Risco residual Baixíssimo (re-identificação) Médio (depende da segurança da chave)

Técnicas

Pseudonimização

  • Tokenização determinística: substitui PII por token; mapping mantido em vault separado
  • Hashing com salt (cuidado — sem salt aleatório, é reversível por força bruta para campos de baixa entropia como CPF)
  • Format-preserving encryption quando o formato precisa ser mantido

Anonimização

  • Generalização: idade exata → faixa (25-34); CEP → cidade
  • Supressão: remoção de campos identificadores
  • Perturbação: adicionar ruído aleatório
  • k-anonymity (k ≥ 5): cada combinação de quasi-identificadores aparece em ≥ k registros
  • l-diversity, t-closeness: refinamentos para resistir a ataques de inferência
  • Differential privacy (ε pequeno): garantia matemática contra re-identificação

Pipeline típico de analytics

[Postgres prod] → ETL com pseudonimização (token PII) → [Analytics DW] → Generalização + k-anonymity → [Dashboards públicos]

Anti-padrões

  • ❌ "Removemos o nome" e mantemos CPF, e-mail, endereço — não é anonimização
  • ❌ Hash de CPF sem salt único — facilmente reversível (~210 milhões de combinações)
  • ❌ ID interno como "pseudônimo" se está exposto em URLs e logs — não é pseudonimização
  • ❌ Agregação só por região quando dataset tem campos suficientes para inferência cruzada

Implementação

Ver assets/anonymization-recipes.md para receitas concretas em SQL e Python.

Re-identificação — teste obrigatório

Antes de declarar um dataset "anonimizado":

  1. Linkage attack: tente combinar com dataset público
  2. Inference attack: estime atributo sensível a partir dos outros
  3. Singling out: existe registro único na base?

Read the full file on GitHub · 69 lines

Files

What ships with it

1 file 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. 12d ago First seen · 69 lines · 90 tokens per session scan A 92273bc421eb

Subscribe to this mod's changes

lgpd-anonymization is a skill published in the GitHub repository goul4rt/lgpd-skills (52 stars, last pushed 3mo ago), licensed MIT. It adds 90 tokens to every session and 767 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

privacy-policy-stephane-boghossian

A zero-hallucination privacy-policy generator that takes anyone — non-lawyer founder to lawyer — from a guided intake to a publishable, jurisdiction-aware privacy policy. Jurisdiction-first: it detects which laws apply from where your users are, then drafts only the required clauses — GDPR/EU + UK, US (CCPA/CPRA, 20…

lawve-ai/awesome-legal-skills · 171 tokens

brasil-legal-skills

Use quando o usuário tiver dúvidas ou precisar de orientação sobre direito brasileiro, declaração de IRPF passo a passo, tributário (Simples Nacional, MEI, CNPJ, CNAE, Reforma Tributária 2026, CBS/IBS), trabalhista (CLT, rescisão, aposentadoria, INSS, CLT vs PJ), societário (abertura de empresa, conflitos entre…

AlissonSantos1/brasil-legal-skills · 159 tokens

digital-brasil

Use para LGPD, proteção de dados, Marco Civil da Internet, cookies, termos de uso, privacidade em apps e sites, direitos digitais, vazamento de dados, adequação LGPD. Ativa em "LGPD", "proteção de dados", "política de privacidade", "vazamento de dados", "cookies", "Marco Civil", "direito ao esquecimento".

AlissonSantos1/brasil-legal-skills · 85 tokens

privacy-compliance

Comprehensive global privacy compliance agent skill covering GDPR, CCPA/CPRA, HIPAA Privacy Rule, EU AI Act, LGPD, cross-border data transfer mechanisms (SCCs, BCRs, EU-US DPF), PII identification and classification, data minimization, consent management, privacy-by-design patterns, DPIA workflows, data subject access…

JPeetz/agent-skills · 132 tokens

compliance-review

Compliance & Governance Review: Reviews systems for regulatory compliance — SOC2, HIPAA, PCI-DSS, ISO 27001, LGPD/GDPR. Covers access control, audit logging, encryption, data retention, incident response, and compliance documentation. Use when the user mentions SOC2, HIPAA, PCI, ISO 27001, LGPD, GDPR, compliance…

camilooscargbaptista/cto-toolkit · 103 tokens

analise-juridica-br

Skill para análise estruturada de contratos brasileiros como apoio a advogado. Acionar quando usuário pedir para "analisar contrato", "revisar cláusulas", "due diligence contratual" ou similar, em contexto de Direito brasileiro. NÃO emite parecer jurídico — produz insumo qualificado para advogado.

falercia/deep-claude · 67 tokens