gdpr-anonymizer

Redact PII from text streams before analysis workflows.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/FolahanWilliams/decision-intel --skill gdpr-anonymizer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gdpr-anonymizer
Source: https://github.com/FolahanWilliams/decision-intel/tree/main/agent-config/skills/gdpr-anonymizer
Command: npx skills add https://github.com/FolahanWilliams/decision-intel --skill gdpr-anonymizer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill acts as a pre-processing layer to ensure Data Minimisation by identifying and masking PII before content is sent to analysis agents, reducing privacy risk.

Core Features & Use Cases

  • PII Identification: Detects Names, Email Addresses, Phone Numbers, specific Locations, Financial Account Numbers, and protected characteristics when irrelevant to the audit.
  • Masking/Redaction: Replaces entities with deterministic tokens (e.g., [PERSON_1], [EMAIL_1], [LOCATION_OFFICE]) and maintains consistency across the document.
  • Verification: Ensures no raw PII remains in the cleaned text and confirms proper handoff to bias_detective or noise_judge.

Quick Start

Apply the gdpr-anonymizer immediately after ingestion to mask PII in the incoming content before routing it to analysis agents.

Frequently Asked Questions about gdpr-anonymizer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I mask PII in text streams before data analysis?

PII redaction works by identifying sensitive entities like names, emails, and financial account numbers in a text stream. It replaces these entities with deterministic tokens like [PERSON_1] or [EMAIL_1] to ensure data minimisation before routing content to analysis agents.

How do I ensure GDPR anonymisation consistency across a document?

You can ensure anonymisation consistency by replacing identified entities with deterministic tokens that remain uniform throughout the document. This masking approach maintains consistent references like [LOCATION_OFFICE] so downstream analysis agents process structured, non-sensitive data.

When do I need to apply PII redaction in multi-agent ingestion workflows?

You need to apply PII redaction immediately after data ingestion and before routing content to analysis agents. This ensures raw PII is removed early, protecting privacy before multi-agent workflows process the text for bias detection or noise evaluation.

What types of personal data are typically identified for data minimisation?

Data minimisation typically targets names, email addresses, phone numbers, specific locations, and financial account numbers. It also identifies protected characteristics when they are irrelevant to the audit, ensuring no raw PII remains in the cleaned text.

How do I verify that no raw PII remains after anonymisation?

You verify anonymisation by running a verification step that checks the cleaned text for residual sensitive data. This process ensures no raw PII remains and confirms proper handoff of the anonymised content to downstream analysis components like bias detectors.