EVERY FACT, ANCHORED
A case intelligence workspace for litigation teams and their customers.
Every fact and contradiction anchored to the document it came from. The chronology, who said what about whom, where two accounts disagree, and what is still missing to prove each issue, each with the document, the page and the quotation behind it.
14-DAY TRIAL · NO CARD · UNLIMITED USERS
Every answer cites the document it came from.
Nothing here floats free of its source. A finding carries the sentence it was drawn from, the page that sentence sits on, the file, and the checksum of that file. One click opens the document on that page with the passage highlighted. A quotation is kept in the original language, because a translated quotation stops being a quotation.
That is also the defence against a confident wrong answer. Every draft and every chat answer is checked against the records before you see it: the document exists, the page exists, the quotation is really there. An assertion with no document behind it has nowhere to hide.
You have read the file. Finding it again is the problem.
The first pass through a case file is work only a lawyer can do. The second pass is not. It is looking for a message you read three weeks ago, somewhere in nine hundred e-mails, while the hearing gets closer.
The chronology goes stale
Disclosure arrives in tranches. A chronology built in March is wrong by May, and rebuilding it by hand is a week nobody has.
Scans are invisible
A served bundle that arrived as images cannot be searched, which means it cannot be relied on during preparation.
Two accounts, never compared
The statement and the contemporaneous documents disagree, but putting every pair side by side is not humanly possible at volume.
Know it, find it, prove it, walk in ready.
Advanced language models read every page once: OCR, transcription, dates, people, claims, and the comparison of every claim against the rest of the file. The reading a paralegal would bill for happens in the background, so the hours go on judgement rather than on looking things up.
Know what happened
Dates, people and events come out of every document and onto one timeline. A profile for every person shows what they said and what was said about them. Each entry names its document and its page.
READ MORE →
Find where the accounts disagree
Statements are broken into claims and each claim is compared with the whole file. A conflicting pair is shown with both quotations, and it stays a suggestion until a lawyer confirms it.
READ MORE →
Prove it
A proof chart puts each issue next to the exhibits that carry it and shows where nothing does. Bundles, chronologies and motions export with the exhibit numbers you set, and every AI draft passes a citation check.
READ MORE →
Walk into the hearing ready
A dossier of everything the file holds about one witness, an examination outline built from it, and every deadline computed from the provision and the date of service.
READ MORE →
ASK THE FILE
A question in plain words, an answer with the page numbers.
The assistant has tools on the records and uses them before it writes: full-text and meaning search, the text of scans and recordings, whole documents read window by window, mail threads in order, the timeline, the claims by speaker, the contradictions and whether a lawyer confirmed them, the profile of a person. When a question needs a number, it runs a read-only query on the case tables. Each tool it reaches for shows up on screen while the answer streams.
Every document in the answer is a link that opens it on the cited page, and a line under the message says how many of the citations were verified against the file. Ask it to file a document into a matter or to leave a note for the team, and it does that too. Nothing else in the file can be changed from the chat.
EVERYTHING THE ASSISTANT CAN DO →TEN QUESTIONS TO START FROM
- Summarise the case
- What did [person] say about [topic]?
- Who is [name]?
- What happened in [period]?
- Timeline of [topic]
- Contradictions about [topic]
- Evidence for [assertion]
- Correspondence with [person]
- Document [ID] in full
- Search the attachments
From folder to chronology in three steps.
Three steps. Nothing to configure, no taxonomy to design first.
-
01
Drop in the file
An archive, a folder, a mailbox export, or mail forwarded to the address the workspace gives you. Formats are recognised by content, so a badly named file still lands where it belongs. Duplicates are caught on checksum before anything runs, and the import screen shows the page count before you commit.
-
02
It reads every page once
Scans go through OCR to the last page, recordings through transcription. Dates, people, organisations and claims come out of the text, and each claim is compared with the rest of the file. It runs in the background, in a queue.
-
03
You work the record
Timeline, profiles, contradictions, the proof chart, deadlines. Or ask the assistant, which searches and reads the file before it answers and cites the page.
What it will not do
- It does not give legal advice, and it does not decide what a document means. That stays with the lawyer.
- It does not file anything with a court.
- It is not an e-discovery platform. There is no privilege review and no review batching.
- An AI finding is a suggestion until a lawyer confirms it, and every generated document opens with the words "working draft".
- Statutory holidays are not applied to deadlines. The provision and the arithmetic are shown, so a holiday end is yours to correct.
That is the boundary that keeps every answer checkable.
Priced by the page, not by the seat.
A plan is a number of case files and a monthly pool of pages of analysis. Count a page as a minute of careful reading, and the smallest pool is 42 hours of it a month. Add the whole team; a seat costs nothing. No quote, no onboarding project, and no credits to run out of.
SOLO
$79/mo
1 CASE · 2,500 PAGES / MO
One case file, read to the last page.
PRACTICE
$299/mo
5 CASES · 10,000 PAGES / MO
For a practice running several client files at once.
FIRM
$799/mo
UNLIMITED CASES · 30,000 PAGES / MO
For a firm that opens a new file every week.
ENTERPRISE
Talk to us
UNLIMITED CASES · PAGES SET WITH YOU
Your own installation, your own AI keys, your own rules.
- A PAGE
- One page of a PDF or scan, about 400 words of e-mail or Word, or a minute of a recording. Counted once, when the file first enters the case.
- RUN OUT MID-CASE?
- A pack of 5,000 pages costs $129 and stays valid for 12 months. Files that arrive early wait, stored, and start on their own.
- NEVER METERED
- Users · storage · chat and search · exports and reports · keeping the file. None of it is on the meter, on any plan.
30-day money-back guarantee on your first payment.
Nothing goes in a filing unverified.
- ONE CASE FILE PER WORKSPACE
- Each workspace has its own content database and its own file prefix. A query in one cannot reach another.
- EVIDENCE IS IMMUTABLE
- Once a document enters the file, its text, its path and its checksum cannot be overwritten.
- THE LAWYER DECIDES
- AI drafts are labelled as drafts and marked review-required. They are a starting point, never an output.
- SOURCES STAY PRIVATE
- Files are streamed through the application. There are no public URLs to case material.
- EVERY AI TEXT CITE-CHECKED
- A draft or a chat answer is checked against the records before you see it: the document, the page, the quotation. What cannot be verified is reported, never quietly repaired.
- A FINDING IS A SUGGESTION
- A contradiction the model found stays a suggestion until a lawyer confirms it, and a dismissed one never comes back. Only confirmed findings reach a draft as facts.
Questions we get asked
What does Legalnaut actually do?
It reads a case file once and keeps it navigable. You get a chronology of events built from the documents themselves, a profile of every person and organisation showing what they said and what was said about them, and a list of places where two documents contradict each other. Every one of those points at the document it came from.
How is this different from searching my documents?
Search finds a document. Legalnaut finds the sentence, then shows you what it conflicts with. The unit of work is a claim made in a document rather than the file the document lives in, which is what makes comparison across a whole case file possible.
What file formats can it read?
PDF, Word, Excel, PowerPoint and their OpenDocument equivalents, RTF, plain text, Markdown, CSV, HTML, e-mail as EML, MBOX and Outlook MSG with attachments, archives, images, and audio and video recordings. Type is recognised by content rather than by file extension, so a badly named document still lands where it belongs. Scans go through OCR on every page, chat screenshots are read as conversations, and recordings are transcribed.
How long does the first import take?
It runs in the background, in a queue, and you can watch the log while it does. Each document is processed as its own job, so a failure on one does not take the rest with it. The page count is shown before the import starts, and a page is counted once, when the file first enters the case.
Does it give legal advice?
No, and it is not built to. It finds and organises what is in your documents. Deciding what a fact means, what to plead and what to do next is the part that stays with a lawyer. Anything drafted with AI is labelled a draft and marked review-required.
What stops it inventing an answer?
Every answer carries the documents it was built from, with their checksums. An assertion without a document behind it is visible immediately, which is the only defence that actually works. Courts have sanctioned lawyers over invented citations, so this is not a theoretical concern.
Is my case file used to train an AI model?
No. Our agreements with the model providers prohibit training on customer content.
Can two matters share a workspace?
Yes. A workspace holds one case file, and matters inside it group documents for a particular thread of work without copying them. What a workspace does not do is mix two clients: for that you open separate workspaces, which is what the plans with several cases are for, and nothing is queryable across two of them.
Navigate the case, not the pile.
Start with one case file and 14 days. No card, nothing to uninstall, and 30 days of money back once you do pay.
See the plans