data-mask

Mask sensitive numerical data in text and supported documents.

195|12|Updated Jul 19, 2026
One-click install
npx skills add https://github.com/ZJU-REAL/HugAgentOS --skill data-mask
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: data-mask
Source: https://github.com/ZJU-REAL/HugAgentOS/tree/main/src/backend/skill_bundles/marketplace/data-mask
Command: npx skills add https://github.com/ZJU-REAL/HugAgentOS --skill data-mask

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pandas, numpy, openpyxl, python-docx, python-pptx, PyPDF2, fpdf2, Pillow, pytesseract, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill removes or obscures sensitive numerical data from text, spreadsheets, office documents, PDFs, web pages, and images before sharing them externally or sending them to AI systems.

Core Features & Use Cases

  • Four masking strategies: Replace numbers with deterministic hash values, placeholders, blurred values, or approximate ranges.
  • Multi-format processing: Handle TXT, Markdown, CSV, Excel, Word, PowerPoint, PDF, HTML, JSON, XML, YAML, and image files with OCR support where applicable.
  • Controlled and reversible workflows: Select spreadsheet columns, skip phone numbers, process files in batches, preserve common document structures, and restore hash- or placeholder-masked data with a mapping file.
  • Use Case: Prepare a sales report for external AI analysis by masking revenue and percentage figures while preserving useful trends, then retain a protected mapping file for authorized restoration.

Quick Start

Use the data-mask skill to mask all sensitive numbers in the attached file with the default hash strategy and save the result as a masked copy.

Frequently Asked Questions about data-mask

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

FAQPage Schema
How do I mask sensitive numbers in Excel and PDF files before sharing?

To mask sensitive numbers in Excel and PDF files, apply hash, placeholder, blur, or range strategies. This obscures numerical data while preserving document structure, supporting spreadsheet column control and optional reversible mapping files.

Can I extract and mask numerical data from images using OCR?

Yes, you can mask numerical data in images using integrated OCR extraction. Tesseract and Pillow dependencies enable text extraction from images, allowing sensitive numbers to be masked before external sharing or AI analysis.

What is the best way to prepare documents with PII for AI data preparation?

The best way to prepare documents for AI data preparation is applying range or blur masking strategies. This obscures exact sensitive numerical values while preserving general data trends required for accurate AI analysis.

Does data masking support batch processing for multiple office documents?

Yes, data masking supports batch processing workflows for multiple office documents. You can process TXT, CSV, Word, PowerPoint, and PDF files in batches while applying configurable numeric privacy protection strategies.

How do I reverse masked numbers back to the original values?

To reverse masked numbers back to their original values, use the protected mapping file generated during hash or placeholder masking. This authorized restoration file securely links masked data to original numerical values.

What are the limitations of using blurred values for data masking?

A limitation of using blurred values for data masking is that exact numerical precision is lost. Blurring replaces sensitive numbers with approximate values, making it unsuitable when applications require exact figures for calculations.