Features

Everything Anomicore does, in plain terms

From detection to review to the final audit log, every step runs on your own machine and every redaction is a decision you make — not one that happens to you.

Detection

Finds the details that identify someone or something

Anomicore combines named-entity recognition with pattern-based recognizers, tuned against real legal, financial, and healthcare documents rather than generic text.

Person names Organizations Locations & addresses Dates Money amounts Percentages Quantities Social Security numbers Phone numbers Email addresses

Exact category coverage depends on your Anomicore edition and continues to expand — see the FAQ for how detection updates are delivered.

File formats

Works with the files you already have

Word

.docx documents, including multi-occurrence replace and undo for every redaction.

PDF

Native text extraction plus OCR for scanned or embedded-image pages, so nothing hides in a picture.

Excel

Cell-by-cell detection and redaction across .xlsx workbooks.

Images & text

Scanned images via OCR, plus plain-text files, for the odds and ends the other formats don't cover.

Review workflow

You stay in the loop on every decision

Automated detection gets you most of the way, but the final call on an ambiguous name or a borderline amount should be yours — not a black box's.

Candidate-by-candidate review

Replace, decline, or replace every occurrence at once — your choice, per item, with a running summary of what's decided.

Live document preview

See the exact text about to be redacted, highlighted in context inside the real document — not just a bare string in a list.

Undo anything, any time

Change your mind mid-review — every decision can be revisited and reversed before you finish.

Custom replacement values

Accept the suggested anonymized value, or type your own — "Person 1" or a value that fits your own document style.

Blackout or pseudonym mode

Choose a solid redaction box, or a readable stand-in like "Person 1" and "Amount 1" — whichever fits the document's purpose.

Category override

Reclassify a detected entity on the spot — if Anomicore calls something an Organization and it's really a Location, fix it inline.

Output & audit

A finished file, and a record of exactly what changed

Clean redacted output

A new file in the same format as your source, ready to share, with the original left untouched.

JSON + Excel redaction log

Every redaction run produces a machine-readable and a human-readable log — what was found, what category, and what was done about it.

Optional original-text record

Choose whether the log includes the original sensitive text for audit purposes, or only what category was redacted.

Want the technical detail on how your data is handled?

See exactly what runs locally, what (if anything) ever reaches a server, and how licensing will work.