INTERACTIVE SPECIMEN:Click any highlighted item to switch between [TOKEN] & original
You want AI’s help with the heavy reading — without handing over your clients’ private lives. Redactorr strips names, Medicare numbers, TFNs, and account details right in your browser, so you can use any AI or LLM with complete confidence.
SECTION 02 — THE METHODORIGINAL NEVER LEAVES THIS MACHINE
How it works
Four steps. You keep the final say.
Redactorr does the heavy lifting — spotting every name, number, and address in seconds — while you stay in charge of what gets replaced and what stays readable.
01 · OPEN IT ON YOUR DEVICE01 / 04
FILE // harbour-mutual-may-demo.pdfSCANNING...
Account holderSarah Mitchell
BSB063-241
Account number1094 7823
02 May · Carlton Property Co. · Rent · −A$2,200
[SCAN: IN-BROWSER]
Drop in a contract, referral letter, or bank statement and Redactorr reads it right on your machine. Names, Medicare numbers, TFNs, and account details are highlighted where they sit — without uploading a single page.
PROCESSED ON YOUR COMPUTER · 0 UPLOADS DURING REDACTION
AUTO-APPROVE THRESHOLD (>95%) ACTIVE2 OF 2 CLEARED
02 May · Carlton Property Co. · Rent · −A$2,200
When Redactorr spots an unmistakable identifier — like a real Australian TFN, ABN, or Medicare card — it marks it ready automatically. When a name or reference could go either way, it pauses and asks you, so you stay in control without reading every line until your eyes blur.
SMART PRIORITISATION · YOU KEEP THE FINAL SAY
03 · PRIVATE DETAILS BECOME READABLE TOKENS03 / 04
FILE // harbour-mutual-may-demo.pdfTOKENISED
Account holderSarah Mitchell[PERSON_1]
BSB063-241[BSB_1]
Account number1094 7823[ACCOUNT_1]
02 May · Carlton Property Co. · Rent · −A$2,200
Black bars ruin a document’s logic — an AI cannot tell whether three black boxes are the same client or three different people. Redactorr swaps private details for clear placeholders like [PERSON_1], so summaries and analysis still make complete sense.
MEANING SURVIVES — AI AND COLLEAGUES CAN STILL FOLLOW THE STORY
04 · ONLY THE CLEAN COPY TRAVELS04 / 04
FILE // harbour-mutual-may-demo.pdfCLEARED
Account holderSarah Mitchell[PERSON_1]
BSB063-241[BSB_1]
Account number1094 7823[ACCOUNT_1]
02 May · Carlton Property Co. · Rent · −A$2,200
RELEASED
What you hand to AI, save to your team Vault, or share with external counsel is the clean working copy. Your original file and the Redaction Map stay safely behind with you.
ORIGINAL + REDACTION KEY STAY BEHIND · YOU CHOOSE WHAT MOVES
Any format. Same discipline.
The engine is purpose-built for the complexity of these formats.
Documents
PDFDOCXRTFODTTXT
Spatial OCR and form field analysis
Track changes, headers, and footnotes preserved
Embedded objects and deep metadata parsed
Layout and structure survive redaction intact
Spreadsheets
XLSXCSV
Multi-sheet workbook traversal
Formula-resolved cell values processed
Named ranges and complex table structures supported
Strict schema adherence for downstream pipelines
Presentations
PPTX
Slide master and layout templates scanned
Speaker notes and hidden text extracted
Embedded diagrams and shapes traversed
OCR
Images & Scans
PNGJPEGTIFFWEBP
Local on-device OCR — no cloud dependency
Automatic skew correction and contrast boosting
Handwriting and low-resolution artefact handling
{
"key": [TOKEN]
}
Code & Data
JSONXMLHTMLMD
Nested keys, arrays, and deeply nested objects
Code blocks, HTML tags, and front matter handled
Output guarantees exact schema parity (it still parses)
Secrets & Keys
KEYSTOKENSCERTS
API keys, bearer tokens, and connection strings detected
Identified inside any of the supported formats above
Flagged as candidates — nothing auto-deleted
SECTION 03 — THE ANALYSISSTRUCTURE SURVIVES REDACTION
Working with AI
Now let AI do the heavy reading.
Hand your AI the clean working copy and get your analysis back in seconds. Your client’s real details never left your desk.
You: Summarise this statement. Anything unusual about the rent? AI: [PERSON_1] paid A$2,200 rent on 02 May — the largest outgoing this month and up from the prior period. Subscriptions are steady; the closing balance is A$6,116. You: Do we hold earlier statements for [PERSON_1]? AI: Two — March and April, both on [ACCOUNT_1]. Rent was A$2,050 in each, so May is the first increase: up A$150, with no other recurring change.
Summarise this statement. Anything unusual about the rent?
AI · ANSWER
paid A$2,200 rent on 02 May — the largest outgoing this month and up from the prior period. Subscriptions are steady; the closing balance is A$6,116.
YOU · QUESTION
Do we hold earlier statements for ?
AI · ANSWER
Two — March and April, both on . Rent was A$2,050 in each, so May is the first increase: up A$150, with no other recurring change.
WHAT AI ACTUALLY RECEIVES · MEANING INTACTPRIVATE DETAILS WITHHELD
Account holder BSB · Account 02 May · Carlton Property Co. · Rent · −A$2,200.00 12 May · Stream Suite bundle · Subscriptions · −A$131.00 24 May · Docklands Electronics · card · −A$1,249.00 Closing balance · A$6,116.00
YOUR LOCAL REDACTION MAP // KEPT ON YOUR DEVICEHOVER OR CLICK ANY [TOKEN]
Click or hover any placeholder below to see the real value kept safely on your device:
Why Placeholders Beat Black Bars
AI doesn't need your client's real name or bank account to understand a document. By swapping private details for consistent labels like [ACCOUNT_1], your AI can still follow who paid whom, spot risks, and draft accurate summaries — without ever seeing the confidential information underneath.
SECTION 04 — THE BOUNDARYWHAT STAYS ON YOUR MACHINE, WHAT CAN MOVE
Your privacy boundary
What stays on your computer. What you choose to share.
Nothing leaves your desk without your say-so. Opening and redacting a document happens entirely on your device — it uploads nothing anywhere. Sharing or sending to AI is always a separate step that you control, and it only ever carries the clean working copy.
ON YOUR COMPUTER · NEVER UPLOADED
Your original documentClient names, account numbers, and private records — kept strictly on your machine.
The Redaction MapThe private key that links each placeholder back to its real name. Kept separate from the copy.
STAYS WITH YOU · 0 UPLOADS
→
THE ONLY GATEYOUREVIEW + APPROVE
→
READY TO USE · ONLY WHEN YOU CHOOSE
The clean working copyReadable placeholders instead of real names — safe for AI, colleagues, and external review.
VAULT · saved for your teamANY AI OR LLM · with your consentSECURE SHARE · expiring link
ONLY WHEN YOU SEND IT
UPLOADS DURING REDACTION: 0RAW DATA SENT EXTERNALLY: NONEPROCESSING: 100% ON YOUR DEVICE
Basic tools only catch obvious emails and phone numbers, and miss the client names buried in a table or annexure. Every full scan runs our 15-stage AEGIS engine: two preparation stages (L01 RECON and L02 SENTINEL), twelve domain specialists (L03–L14), and a final reviewer (L15 WARDEN) that holds anything uncertain for your call.
15WARDEN · final quality check: resolves overlapping matches and holds anything uncertain for youHOLD
→YOU · approve, edit, reject, retag. The only release authority on the pipeline.GATE
Global tools treat Australian identifiers as edge cases. We treat them as the default.
Every AU government ID, healthcare number and financial identifier ships built in, not as an add-on.
38AU government identifiers
TFN, ABN, ACN, Medicare, BSB, DVA, WWCC, Director ID, ImmiCard, Centrelink CRN, IHI, NDIS, PBS, USI and more: the identifiers Australian documents actually carry, built into one engine.
8Official format verification
Real Australian TFNs, Medicare cards, and ABNs follow strict national rules. Redactorr checks those rules automatically, so ordinary invoice or page numbers are left alone instead of cluttering your review.
MAPPEDAustralian laws + regulations
Privacy Act 1988, Healthcare Identifiers Act 2010, My Health Records Act 2012, NDIS Act 2013, Migration Act 1958 and more: entity types are linked to their governing legislation where applicable.
BSBBank directory checks
Every BSB is validated against an institution registry that ships inside the engine, including Big 4, Macquarie, Bendigo, ING, Suncorp, BOQ, St.George, Bankwest and more. No server call needed.
LOCALPrivacy Act 1988 · data boundary
Detection and redaction run in your browser and original values stay local. The boundary is part of the coverage, not a separate promise. Copies move on only when you send them.
DOMAINIndustry configurations
Construction, healthcare, legal, insurance, immigration, aged care, government, real estate and more: each configuration is tuned to that industry’s documents and the identifiers that matter in them.
Trust isn’t a paid feature.
Whether you are testing a snippet in the Free 20-pattern checker or running a 200-page brief through the 15-layer AEGIS engine on Individual User or Enterprise plans, the zero-upload browser security boundary is identical on every plan.
06SEPARATION · your Redaction Map is kept strictly separate from the clean working copy at all times
15AEGIS layers per file
100+Entity types detected
38AU identifiers built in
15+Document formats
0Uploads during redaction
It started with a simple question: why does every AI tool demand your raw data?
3,000+ detection patterns · every edge case hunted downRead our story →
SECTION 06 — THE PACKSYOUR INDUSTRY, DETECTED
The packs
27 Industry packs. One dedicated engine each.
From legal discovery and clinical notes to FOI and tax audits. Select a sector below to inspect its dedicated entity detectors and real-world AI workflow — or explore all 27 industry packs.
Court filings, briefs, and client matters contain privileged information. One mishandled document ends careers.
Targeted Entity Detectors
Client names
Case numbers
Court file references
Witness details
Legal privilege markers
Solicitor details
Settlement amounts
Sanitised AI 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.
SHOWING PACK 01 OF 27 AVAILABLE SECTOR ENGINES
SECTION 07 — THE VAULTSTORED, HELD, GOVERNED APART
The vault + team workflows
Keep the clean copy moving. Keep the original private.
Redacting a file is only step one. Redactorr gives your team a safe place to store clean working copies, share them via expiring links, and use them across any AI tool — while your clients' original names and numbers stay locked away.
01 · REVIEW
Check the original on your computer
Redactorr highlights every sensitive name, number, and address right in your browser so you can approve, adjust, or skip any item before saving.
On your device · 0 uploads
02 · SAVE
Keep a clean working copy
Once approved, your clean copy is ready for everyday work — whether you are dropping it into AI, storing it in your team Vault, or briefing counsel.
Clean working copy · Ready to use
03 · PROTECT
Keep the key separate
The Redaction Map — which links each placeholder back to a real name — is kept separate from the working copy, so private details never travel by accident.
Key kept apart · Protected
04 · RECORD
Know who saw what, and when
Every share link, expiry date, and review certificate is logged clearly so your practice always has a clean audit trail for clients and regulators.
Your team shouldn't have to choose between modern AI productivity and client confidentiality. Redactorr lets you hand a clean, readable copy to any AI or LLM while the real names and your Redaction Map stay safely behind on your device.
Only the clean working copy ever reaches your AI toolYour Redaction Map stays behind on your device unless you choose to restore namesEvery summary and answer keeps full document context and structure
WORKFLOW // SAFE AI HANDOFF[ORIGINALS STAY LOCAL]
STEP 01 // CLEAN COPYTOKENISED
Patient: [PERSON_1]
Account: [MEDICARE_1]
Full story and relationships preserved.
SEND→
STEP 02 // AI RESPONSESAFE
Summary ready
Zero private details shared
AI answers accurately using placeholders.
KEPT WITH YOUYour original document and Redaction Map never leave your computer.
Custom Rules
Teach Redactorr your firm’s own codes.
Every practice has its own internal matter numbers, project codenames, or staff IDs. Add your own custom formats or word lists in seconds, and Redactorr will catch and replace them automatically alongside standard personal details.
Handover notes for [PROJECT_CODE_1] and [PROJECT_CODE_2] cleared for partner review.
SECTION 08 — THE PRICE100% LOCAL ISOLATION, EVERY PLAN
TAILORED LICENSING
Tailored plans. Quoted for your exact seat count.
Every licence includes our full 15-layer client-side AEGIS engine, Custom Practice Packs, and 100% local in-browser isolation, with custom quotes calibrated to your number of users.
INDIVIDUAL AND PRACTITIONER
Individual User
Custom Quote
1 to 10 Users · Calibrated to your seat count
For solo lawyers, clinicians, accountants, consultants, and boutique practices requiring full local document redaction and Custom Practice Packs.
Number of Users:1 User (Solo)
Full 15-Layer AEGIS Engine (3,000+ Rules)Custom Practice Packs and Dictionaries includedUnlimited multi-page PDF, Word, Excel and OCR100% local in-browser isolation (zero server egress)Reversible Tokens (.TOK), Masking and Forensic BlackoutSigned SHA-256 audit certificate and Encrypted Vault
ENTERPRISE AND ORGANISATION
Enterprise
Custom Quote
10 to 500+ Users · Scaled by organisation seats
Full client-side redaction engine and Custom Practice Packs for multi-team practices and organisations, with dedicated infrastructure, Enterprise Single Sign-On (SSO), advanced auditing, and custom SLAs.
Enterprise Users:25 Users
Full 15-Layer AEGIS Engine (3,000+ Rules)Custom Practice Packs and Dictionaries includedDedicated Delivery Pipeline: Isolated release ring and version pinningDedicated Environment, Private VPC/Air-Gapped and Custom Engine Deployment as a ServiceEnterprise SSO (SAML/OIDC), RBAC and admin-locked policy enforcementAdvanced Compliance Auditing (SIEM Export) and Sovereign Keys (BYOK)Dedicated Enterprise Queues, Custom SLA and MSA/DPA infosec onboarding
Capability Comparison and Onboarding
Compare Individual User vs. Enterprise Licensing
Review the full side-by-side capability matrix or submit your user count directly to receive your custom quote and onboarding link.
SECTION 09 — THE BRIEFINGCISO, LEGAL & PROCUREMENT ANSWERS
Common questions
Straight answers before you decide.
The practical questions Managing Partners, General Counsel, Practice Managers, and Security Leads ask us before bringing Redactorr into their daily workflow.
Q01How can our security team verify that zero bytes of our original document leave the browser?+
You do not have to take our word for it: your IT or security lead can verify it in thirty seconds. Open your browser’s Developer Tools (⌥⌘I on Mac or F12 on Windows), select the Network tab, and drop a document into Redactorr, or turn off Wi-Fi entirely after the page loads. Every stage of detection and redaction runs locally on your computer. Outbound network traffic during redaction is zero.
Q02How does Redactorr help with the Australian Privacy Act (APP 8 and APP 11) and APRA CPS 234?+
Under Australian Privacy Principle 8 (Cross-Border Disclosure), sending raw personal information to an overseas AI provider can expose your organisation to liability if that data is mishandled. Because Redactorr replaces names, Medicare numbers, and TFNs on your own device before you copy anything into an AI tool, your clients’ personal details never cross your border. Every paid scan also produces a signed audit certificate for your compliance records.
Q03What is the difference between the Free PII Checker and the full 15-Stage AEGIS Engine?+
Every plan shares the exact same local, zero-upload privacy boundary. The Free PII Checker covers 20 core patterns for quick text snippets and single-page previews (up to 20 scans a day). Individual User and Enterprise plans unlock the full 15-stage AEGIS engine, covering 3,000+ detection rules, Custom Practice Packs, official Australian government and healthcare ID verification, multi-page PDF, Word, Excel, and scanned document support, plus clean exports with audit certificates.
Q04Why replace private details with placeholders like [PERSON_1] instead of black bars?+
Black bars destroy the story inside a document: an AI or colleague cannot tell whether three black boxes refer to the same client, a witness, or a bank account. Clear, consistent placeholders like [PERSON_1], [MEDICARE_1], and [ACCOUNT_1] keep the entire narrative intact so any AI or LLM can summarise, compare dates, and answer questions accurately while the real names stay safely on your computer.
Q05What is the difference between Individual User and Enterprise licensing?+
Every document-level capability, including the full 15-layer AEGIS engine, Custom Practice Packs, unlimited multi-page PDF/Word/Excel/OCR support, reversible tokens, SHA-256 audit certificates, and Encrypted Vault, is identical across both SKUs. Both plans are custom quoted based on your number of users, with Enterprise adding a Dedicated Delivery Pipeline (version pinning), Dedicated Environment and Custom Engine Deployment as a Service (including Private VPC/Air-Gapped options), Enterprise SSO (SAML 2.0 / OIDC), Role-Based Access Control (RBAC), Centralised Audit Logs with SIEM export, Sovereign Key Management (BYOK), Dedicated Enterprise Queues, and a contractual Enterprise SLA.
Q06Can our CISO or IT team walk through your architecture with your engineers before we roll out?+
Yes. We protect our proprietary engine on the public web, but for legal practices, healthcare groups, financial institutions, and enterprise buyers, we offer a private Architecture Review Walkthrough with our engineering team alongside our downloadable Executive Security Brief.
Give AI the clean copy. Keep the original private.
Redact on your own computer in seconds. Approve every change. Then use any AI tool or share with your team knowing your clients' private details never left your desk.