supplier-agreement.pdfPRIVATE REVIEW
Privacy layer for document work
Redact locally. Work safely with AI.
Turn sensitive documents into reviewed, redacted working copies for AI, sharing, and knowledge workflows — while the original-value map stays separate.
Platform workflow
Control the copy, map, share, and audit trail.
Redaction is the first trust boundary. Redactorr then helps teams move the reviewed working copy into vault, sharing, AI, agents, and knowledge workflows while the original-value map stays separate and governed.
01
Review the sensitive original
Detection and redaction run in your browser during the review step, so candidate values can be approved, edited, or rejected before export.
02
Create the safe working copy
The reviewed redacted copy becomes the useful document object for vault storage, sharing, AI, and knowledge workflows.
03
Keep the map separate
The original-value map is treated as a separate governed object, not bundled into the default copy or share package.
04
Log who sees what
Sharing, AI context, map access, expiry, and review outcomes become visible product states rather than loose feature claims.
Built for sensitive document workflows across
Healthcare
Legal
Finance
Government
Insurance
Enterprise
TRUST CENTER
Original Values Stay Local During Redaction
Watch Redactorr detect and redact sensitive information inside your browser. Reviewed redacted output can then move to Vault, sharing, or AI workflows when you choose.
SCAN
DETECT
REDACT
CLEAN
Browser-local detection
Browser-local redaction
Raw values stay local by default
AEGIS Detection Pipeline
15 layers for document-aware detection.
Two pre-detection layers prepare the document. Twelve specialist engines run simultaneously. One final gate checks unresolved candidates before output.
INGEST
Pre-detection
Two layers prepare the document before any specialist engine fires.
01RECON
Document understood before it is judged. Nothing placed without context.
02SENTINEL
Routes the document to specialist engines so entities are less likely to be handled by the wrong detector.
COMMANDER
COMMANDER
orchestrates 12 specialist engines
Each engine handles a distinct detection domain. They run in parallel, then converge.
FINAL GATE
15 WARDEN
Review gate before release. Unverified findings stay flagged for human review.
Medical Referral
Specialist Referral Letter
14 / 14 REDACTED
Document type: [DOC_TYPE]
Referring Practitioner: [NAME] — AHPRA: [AHPRA]
Patient DOB: [DOB], Medicare: [MEDICARE], TFN: [TFN]
Address: [ADDRESS] | Phone: [PHONE] | Email: [EMAIL]
Referral to: [ORG]
Physician: [NAME]
Diagnosis: [DIAGNOSIS]
Billing: BSB/Account: [BSB/ACC] | Access token in notes: [TOKEN]
AI-safe document workflow
Give AI the reviewed copy, not the raw original.
AI is downstream from document review. Redactorr can provide reviewed redacted context with source labels while raw values and the original-value map remain withheld by default.
Reviewed safe copy enters the AI workspace
Original-value map is withheld unless policy allows recovery
Outputs keep source lineage and approval state
AI-visible
Reviewed working copy
Redacted context, document structure, and approved source labels.
Output
Summary, answer, task
AI can reason over the safe copy while provenance stays visible.
Withheld by default:raw original values and the original-value map.
Workflow pillars
One safe copy can support more than one job.
A reviewed working copy can move into storage, AI, agents, knowledge, and sharing without carrying the raw original or opening the original-value map by default.
Behind the Engine
Redact Locally. Keep Working.
One Relentless Mission.
It started with a simple question: why does every AI tool demand your raw data? That question became an obsession. 3,000+ detection patterns. industry-specific engines. Every edge case hunted down.
The architecture is the privacy boundary.
Tokenized, Not Blacked Out
Traditional redaction destroys context. Redactorr replaces sensitive data with meaningful tokens that preserve document structure.
Traditional
Dear ████, your salary of ████ has been approved.
Redactorr
Dear [PERSON_NAME_1], your salary of [SALARY_1] has been approved.
Browser-Local Redaction
Detection and redaction run in your browser. Original sensitive values stay local during detection and redaction — whether it's a native PDF, a scanned page, a photo of a paper document, a handwritten form, or a Word file.
Your DocumentBrowser
Redacted OutputBrowser
Server UploadsNONEServer
100+ Entity Types
Names, addresses, Tax File Numbers, ABNs, Medicare numbers, credit card numbers, phone numbers, and hundreds more across industry-specific domains. Built for Australian documents.
Your Rules, Not Ours
Manual redaction for surgical precision. Custom patterns for your business-specific identifiers. The engine proposes. You decide what gets redacted.
✓Select any text and redact manually
✓Build custom patterns for proprietary codes
✓Accept, dismiss, or reclassify detections
✓Retag entities to the correct type
DOMAIN PACKS
Industry Packs. One Dedicated Engine Each.
Enable an industry pack for industry-specific detection. Add custom patterns for your organisation's own identifiers. Manually redact anything else. Your data. Your rules.
Legal & Law Firms Pack
Court filings, briefs, and client matters contain privileged information. One mishandled document ends careers.
WHAT REDACTORR DETECTS
Client names
Case numbers
Court file references
Witness details
Legal privilege markers
Solicitor details
Settlement amounts
REAL-WORLD WORKFLOW
“A paralegal needs AI to review 200 pages of discovery documents. Redactorr strips client names, case references, and privileged communications — the AI analyses argument structure without exposure.”
12 specialised patterns|8 entity types
Explore Pack →
Healthcare & Medical Pack
Patient records contain the most sensitive data any organisation handles. There is zero margin for error.
WHAT REDACTORR DETECTS
Patient names
Medicare numbers
IHI (Individual Healthcare Identifier)
Practitioner registration
Diagnosis codes (ICD-10)
Medication details
Provider numbers
REAL-WORLD WORKFLOW
“A clinical researcher needs AI to summarise 500 patient discharge notes. Redactorr masks every patient identifier, Medicare number, and diagnosis — the AI analyses treatment patterns, not people.”
15 specialised patterns|10 entity types
Explore Pack →
Finance & Banking Pack
Account numbers, BSBs, and transaction records are the highest-value targets for data breaches.
WHAT REDACTORR DETECTS
BSB numbers
Account numbers
Credit card numbers (PAN)
ABN/ACN
Transaction references
SWIFT/BIC codes
Superannuation member IDs
REAL-WORLD WORKFLOW
“A financial analyst needs AI to identify trends across 12 months of transaction data. Redactorr strips account numbers, BSBs, and customer names — the AI sees patterns, not identities.”
14 specialised patterns|9 entity types
Explore Pack →
Education Pack
Student records, ATAR scores, and enrolment data are protected under privacy legislation. Schools can’t afford leaks.
WHAT REDACTORR DETECTS
Student IDs
ATAR scores
USI (Unique Student Identifier)
Enrolment numbers
Parent/guardian names
Teacher registration numbers
Academic transcripts
REAL-WORLD WORKFLOW
“A university administrator needs AI to analyse enrolment trends across 5 years of records. Redactorr removes student names, IDs, and ATAR scores — the AI works with demographics, not individuals.”
8 specialised patterns|7 entity types
Explore Pack →1 / 27 — Legal & Law Firms Pack
Work with the documents teams already use.
PDFs, Word files, spreadsheets, images, and plain text can move through the same review-first redaction workflow.
ALSO SUPPORTSCSV, RTF, HTML, JSON, XML, PNG, JPEG, TIFF, ODT, PPTX, TXT + more
15+
document formats
100+
entity types detected
15
AEGIS layers per file
0
server uploads
Security
Trust isn't a paid feature.
Same infrastructure on every plan.
Local
original values during redaction
Browser
sensitive-value detection and redaction
Review
before redacted output moves forward
Authentication & Access
Workspace access is gated by authenticated accounts, role-aware permissions, and session controls.
Data Protection
Detection and redaction run in your browser. Original sensitive values stay local during detection and redaction. Processed redaction sessions can be saved to Vault when you choose.
Detection Coverage
Australian document patterns, financial identifiers, health references, and industry-specific entity types are mapped for review before export.
Review History
Review history focuses on processed redaction sessions, accepted detections, exports, and share-ready redacted packages.
Secure Sharing
Passphrase-protected links, automatic expiry, view-count limits, instant revocation. Selective unredaction per recipient.
Detection Integrity
3,000+ detection patterns. Mathematical verification for every structured identifier. PDF vulnerability scanning before processing begins.
Free Forever. Pro When You Need It.
The full detection engine is free, including manual redaction. No trials that expire. Upgrade for unlimited scans, document upload, and custom patterns.
Free
Freeforever
No credit card required. No expiry.
✓5 documents/month
✓All document formats — PDF, Word, Excel, images & more
✓10-page max per document
✓1 AI session/month
✓Mandatory watermark
✓Runs in your browser
✓Community support
Most Popular
Pro
A$99/month
A$82.50/mo billed annually (2 months free)
For professionals who handle sensitive data daily.
Everything in Free, plus:
✓Unlimited documents, all formats
✓Coverage across the 15-layer AEGIS workflow
✓150 AI sessions/month
✓50 custom patterns
✓Audit trail
✓Priority support
Team
A$199/month
3 seats included. Additional seats A$69/seat/mo.
Drag to adjust team size
3 seats
3
5
10
20
50
Everything in Pro, plus:
✓150 AI sessions/seat/month
✓Shared team workflow
✓Shared pattern library
✓Dedicated support
✓Up to 50 seats
Enterprise
Built for teams above 50 seats.
Custom pricing. Bespoke engine configuration. Dedicated support. Custom integrations and deployment planning.