PermanentRedaction is a web-based redaction platform built for legal and professional document workflows where sensitive information must be permanently removed from text-based PDF documents.

Key features and functions include:

Rule-Based Discovery
Users can create keyword, phrase, and regular expression rules to search for sensitive text such as names, dates, IDs, and account numbers across one or multiple PDFs in a single job. Rule sets can be saved, imported, and exported for reuse across jobs.

Manual Redaction Tools
Manual redaction boxes are available for content that cannot be reliably located through text search, including image regions within scanned or image-based PDFs.

Preview and Review Workflow
Before finalizing, users can review discovered text matches and image regions and confirm that intended redactions are positioned correctly prior to permanent application.

Verification Reporting
Completed jobs include a Verification Report intended to confirm that targeted text is absent from the finalized document, generated through a two-pass verification process.

Redaction Index
A confidential Redaction Index is generated for completed jobs, documenting matched values, counts, pages, and rule details for internal review.

Flexible Output Options
Users can export text-preserving redacted PDFs or fully flattened image-only output, depending on filing or sharing requirements.

Multi-Document Job Support
Redaction rules can be applied across multiple PDFs within a single job rather than requiring documents to be merged.

Deterministic, Non-AI Workflow
The platform does not send documents to AI models for interpretation, classification, or redaction decisions. Redaction rules and outcomes are defined and controlled by the user rather than by automated content analysis.