CasePrepd
Case prep that never leaves your computer.
Vocabulary lists, semantic search, and key excerpts from your case documents — processed 100% on your machine. Nothing uploaded. Nothing stored in the cloud.
Built for sensitive court matters, HIPAA, and anything you can't risk sharing.
Download for WindowsBuilt by Noah Collin, a court reporter.
On this page
What You Get
Vocabulary Lists
Names, medical terms, legal jargon, and technical words automatically pulled from your documents. Scored by importance so you prep the terms that matter most. The app learns your preferences over time — the more you use it, the better it gets at flagging what matters to you.
Semantic Search
Ask natural-language questions about your case documents and get answers with source citations. Great for targeted questions across a breadth of documents.
Key Excerpts
Surfaces topical themes from your documents by grouping related passages. A starting point for getting oriented in a large case — not every excerpt will be a hit, but the good ones save real time.
Export vocabulary to Word, PDF, HTML, or plain text. Search results also export to CSV.
Screenshots
1
2
3
4
5
- Manage saved corpora (collections of past job documents) for comparing terms across jobs
- Drop files here or click to add PDFs, Word docs, and text files
- Switch between Document, Vocabulary, Search, and Key Excerpts tabs
- Process all loaded documents with one click
- Export results to clipboard, file, or formatted report
Full text extracted from your PDFs and Word docs, with page-by-page navigation.
Multiple methods work together to pull the names and terms that matter most from your documents. Each term is scored by relevance, and the app learns from your Keep/Skip feedback to surface better results over time — tuned to what court reporters and scopists actually need for job dictionaries.
Ask plain-English questions about your case documents and get answers with source citations. Useful for pulling specific facts from hundreds of pages.
Passages grouped by topic to help you get oriented in a large case. Some excerpts will be more useful than others — think of it as a starting point, not a summary.
A corpus is a saved collection of past job documents. CasePrepd compares new documents against your corpus to highlight terms that are unusual for your line of work — so common legal boilerplate drops to the bottom and case-specific names rise to the top. You can keep several corpora (one for med-mal, one for criminal work, etc.) and the more jobs you add to each, the smarter the scoring gets.
Who Is This For?
- Court reporters prepping vocabulary for upcoming jobs
- Scopists and proofreaders building job dictionaries from case documents
- Anyone handling sealed, confidential, or HIPAA-protected materials
- Anyone wary of sending documents to an AI cloud
If you're a court reporter or scopist, see what makes CasePrepd different from generic AI tools — the name canonicalizer, transcript artifact stripping, role detection, and the other court-reporting-specific logic that's actually in there.
How It Works
CasePrepd is a desktop app, not a web service. You load your documents (PDF, Word, text files, and even scanned images), and everything is processed locally using built-in AI. No internet connection is needed at any point — not during installation, not during use. The app never connects to the internet, period.
Under the hood, CasePrepd uses multiple on-device AI models for search, name recognition, and vocabulary filtering. Everything ships with the installer — nothing is downloaded at runtime.
Built Specifically for Court Reporting
The logic inside CasePrepd was built by a working court reporter, not adapted from a generic AI tool. It knows that "DI LEO 1 Q" and "Di Leo: Objection" are the same speaker. It collapses OCR variants like "Arthur Jenkins" / "Anhur Jenkins" / "Arthur Jenidns" into one canonical entry and shows you which spelling won and why. It strips line numbers, page headers, and title pages automatically. It tags people by role (treating physician, plaintiff, opposing counsel) so you know who matters at a glance.
System Requirements
- Windows only (Windows 10 or 11)
- 16 GB RAM minimum
- 4 GB disk space minimum
Download
v1.0.35 — May 2026
On the next page, click CasePrepdSetup.exe to start the download.
Windows SmartScreen Warning
CasePrepd is not code-signed (code signing certificates cost hundreds of dollars per year, and this is free software). When you run the installer, Windows SmartScreen may display a warning: "Windows protected your PC."
To proceed: click "More info", then click "Run anyway". The source code is fully open and available on GitHub for inspection.
Getting Started
Install
Run the installer. No internet needed — everything is included.
Load Documents
Drag and drop your PDFs, Word docs, or text files into the app.
Process
Click one button. Vocabulary, search, and key excerpts are ready in minutes.
Want the longer tour? See the User Guide for what each tab does, how to use the Keep/Skip thumbs, and reference libraries. The Power User Guide covers the command-line interface and advanced settings.
FAQ
Why is this free?
I'm a court reporter who built this for my own use and wanted to share it with the community. It's open source and always will be.
Is my data really private?
Yes. CasePrepd never connects to the internet. All processing happens on your machine using models that ship with the installer. No data is uploaded, logged, or transmitted anywhere.
Do I need a powerful computer?
You need 16 GB of RAM and about 4 GB of disk space. CasePrepd runs entirely on CPU — no GPU required.
What file types are supported?
PDF, Word (.docx), plain text (.txt), and images. Scanned PDFs are automatically read using built-in text recognition (OCR) — no extra software needed.