seanpedrick-caseseanpedrick-caseCommunity·5 Agent Skills Included

doc_redaction

Redact personal data from PDFs, images, and spreadsheets

Detects and removes personal information from PDFs, images, Word files, and spreadsheets using OCR and entity recognition. Supports local, AWS, and LLM-based detection with review files, coverage checks, and a two-pass visual verification workflow. Eliminates manual page-by-page redaction and helps teams meet privacy and compliance requirements faster.
npx skills add seanpedrick-case/doc_redaction --all -g -y
Available:

Gives the agent project context, setup steps, memory and OCR tuning options, and routes it to the correct redaction skills for first-pass redaction, page review, and reapplying edits.

All Skills in This Repository (5)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install doc_redaction?

Run `npx skills add seanpedrick-case/doc_redaction --all -g -y` in your terminal to install all skills in this suite globally.

What file types can doc_redaction redact?

It redacts PII from PDFs, PNG and JPG images, Word documents, and tabular files like CSV, Excel, and Parquet.

How does the two-pass review workflow work?

Pass 1 uses OCR and CSV review files to verify and fix redaction boxes, then applies them once. An optional Pass 2 uses a vision model to visually check flagged pages for missed handwriting or signatures.

Can I redact documents without cloud services?

Yes. Local OCR engines and spaCy-based detection run entirely on your machine, with AWS Textract, Comprehend, and Bedrock available as optional integrations.

Does doc_redaction work with coding agents?

Yes. It ships SKILL.md playbooks and a FastAPI agent endpoint so agents can run redaction, review pages in parallel, and verify coverage programmatically.

Related Repositories in Legal & Compliance

View All in Legal & Compliance