Works with the providers you already use
OpenAI’s GPT-4o and GPT-4o-mini. Anthropic’s Claude. DeepSeek. Bring the model you trust.
Ogma Library
You have spent years collecting PDF books. Hundreds, maybe thousands — textbooks, biographies, technical references, novels, research papers, business classics. Each one an investment. And right now every single one is sitting in a folder called “Books”, named things like clean_code_final_v2.pdf. Invisible. Unsearchable. Forgotten.
Ogma Library changes that.
Coming soonBuilds are in final testing. Get notified the moment they’re ready →
The problem
Here is the honest truth about PDF book collections: the way most people store them is barely better than not having them at all.
You cannot see what you own without opening each file. You cannot remember what you read or how far you got. You cannot discover what you have that might be exactly what you need right now. When you want something to read, you reach for the same five books you always do — or you buy something new that you might already own.
The books are there. The knowledge is there. The problem is everything around them.
Windows Explorer was built for files, not for libraries. It shows filenames and dates. It has no concept of authors, publishers, or ISBNs. It cannot tell you that you’re on page 247 of something you started three months ago. It cannot look at your collection and tell you what to read next.
The product
Ogma Library turns a folder of cryptic filenames into an ordered, intelligent, beautiful place to read — one that works fast and actually helps you read more.
The first time you see your collection in Ogma Library, something shifts. Your books are no longer a list of filenames — they are physical objects on wooden shelves, covers facing you, spines lit by warm directional light. Hover over a book and it leans forward to greet you. Click it and the detail panel opens. Double-click and you’re reading.
The bookshelf is rendered in real-time 3D using the same graphics engine that powers many of the world’s most sophisticated web experiences — running natively at a smooth sixty frames per second.
And four ways to look: a real-time 3D Shelf, a cover Grid, a sortable List, and a Directory view that maps your library exactly as it sits on disk. Every view is one click away.
Here is something remarkable: Ogma Library knows what your books are.
Set your library folder once and Ogma scans every PDF. It generates a cover thumbnail from the first page, reads any embedded metadata, and — for books with an ISBN — reaches out to Open Library and Google Books to fetch the full record: publisher, year, edition, language, categories, description, and a high-quality cover.
Four hundred PDFs with cryptic filenames become four hundred properly identified books — authors, covers, descriptions, categories — without you touching a single record by hand. When the metadata is correct, you can write it straight back into the PDF, so the file is properly tagged everywhere, forever.
Ogma Library does more than manage your books. It reads them, and it reads them seriously. Built on PDFium, the same engine inside Google Chrome, it handles any PDF at full fidelity: sharp text, crisp images, complex layouts exactly as designed.
Read single-page, two-page spread, or continuous scroll. Fit to width, zoom to a custom percentage, or go full-screen with every chrome element hidden. Everything is a keyboard shortcut away.
And it remembers exactly where you were. Close a book on page 312 and reopen it a week later — it offers to resume from page 312. Progress is tracked as a percentage, a page count, and cumulative time. Highlights, notes, and bookmarks live in your library database and never modify the PDF files themselves.
Type in the search bar and every book whose title, author, description, category, or tag matches lights up — in the bookshelf, non-matching books fade to near-invisible while your results glow.
Advanced filtering narrows by reading status, rating, category, language, year, or any combination. Sort by anything — alphabetical, recently added, last read, highest rated, longest unread. Save filters as named views and switch between them from the sidebar.
Ogma AI
Open the Ogma AI panel and describe what you want to read. Not a title. Not an author. What you want — your intent, your curiosity, your need right now.
Ogma reads your entire catalogue — every title, author, category, and description — and returns twelve books from your own collection that best match what you described. Not generic bestseller lists pulled from the internet. Books you already own, chosen for your stated intent, each with a paragraph explaining exactly why it fits.
It is the difference between owning a library and having a librarian.
OpenAI’s GPT-4o and GPT-4o-mini. Anthropic’s Claude. DeepSeek. Bring the model you trust.
For complete privacy, choose Ollama — a fully local AI. Nothing about your library ever leaves your computer. Not the titles. Not the authors. Nothing. The recommendations are generated by your own hardware.
Your API keys live in your operating system’s secure vault — Windows Credential Manager, or Keychain on macOS. Never stored in a file. Never visible in logs.
Portability
Ogma stores everything — metadata, covers, reading progress, bookmarks, highlights, notes, shelves — in a single SQLite database file. One file. Fully portable.
Take it to your laptop, a new computer, or a colleague’s machine. Copy the database alongside your books, point Ogma at the folder, and your entire library loads instantly. No rescanning. No re-entering metadata. No losing your place.
Books match their records by relative path; if files were reorganised, Ogma falls back to a content hash — so your progress survives even if you renamed every folder.
\\NAS\Books), mapped drives, and macOS SMB mounts (/Volumes/Books).
Virtual shelves
Your operating system organises files by location. Ogma organises books by meaning. Create virtual shelves — named collections independent of where files live on disk.
Put Antifragile, The Black Swan and Fooled by Randomness on a “Nassim Taleb” shelf no matter which subfolders they’re in. Tag freely. Rate one to five stars. Mark books Currently Reading Finished Abandoned. Your library adapts to how you think.
Reading statistics
How many books did you finish this year? How many pages last month? Which subjects do you read most — and which have you been meaning to finish for six months?
Drawn from real reading sessions Ogma tracks every time you open a book — not self-reported, not estimated. Knowledge about your habits is the first step toward better ones.
Who reads with Ogma
Built with intent. Built to last.
Ogma Library is an original product from Chwezi Core Systems — a technology, business and security consultancy based in Kampala, Uganda. A registered Ugandan company serving organisations across East Africa and beyond, Chwezi designs software, builds custom systems, and ships a growing family of commercial products including Maduuka POS, Aqar, Medic8, Academia Pro and Longhorn ERP. Ogma is the newest.
The name comes from Ogma — the Celtic Irish deity credited with inventing the Ogham alphabet, the ancient script carved as notches into standing stones across Ireland and Scotland. Thousands of those stones survive today, readable after fifteen centuries. Your library deserves the same permanence.
It’s built with C# and .NET 8, using Avalonia UI for a native experience on Windows and macOS. There’s no subscription on the core features. Your data is yours — it sits on your machine, in a portable file you control, and it will be there as long as you want it.
Peter Bamuhigire
Founder, Chwezi Core Systems · Kampala, Uganda
Peter leads Chwezi Core Systems, a full-fledged ICT company delivering software, business and security solutions across East Africa — and he builds Ogma Library himself, line by line. Ogma is what happens when a developer who loves books refuses to let his own collection rot in a folder.
Questions
Ogma Library is a PDF reader, a library management system, a personal knowledge base, and an e-learning companion — built to grow into all of them.
Ogma Library is a native PDF library manager and reader for Windows and macOS. It turns a folder of PDF books into an organised, searchable personal library — with automatic covers and metadata, a fast PDFium reader, reading-progress tracking, virtual shelves, statistics, and an AI advisor that recommends books from your own collection. In one app it serves as a personal library management system, a knowledge base, and an e-learning and study companion.
Yes. Ogma includes a serious reader built on PDFium — the same engine inside Google Chrome — with single-page, two-page spread and continuous scroll, fit-to-width and custom zoom, full-screen reading, keyboard shortcuts, and exact page resume so you always pick up where you left off.
Yes. Ogma is built as a library management system for hundreds or thousands of PDFs. It scans your folder, generates covers, fetches bibliographic metadata by ISBN from Open Library and Google Books, and lets you search, filter, sort, tag and organise everything into virtual shelves.
Yes. Organise books by meaning with virtual shelves and tags, rate them, and capture highlights and notes that are stored in your library database — never written into the PDF files. Full-library search finds anything across titles, authors, descriptions, categories and tags.
Yes. Ogma tracks real reading sessions, page counts, cumulative time, completion percentages and streaks — a practical study and e-learning companion for students, researchers and lifelong learners.
OpenAI (GPT-4o and GPT-4o-mini), Anthropic Claude, and DeepSeek — or Ollama for fully local, private AI that runs on your own hardware and never sends your library anywhere. The advisor recommends books from your own collection, with a reason for each pick.
Yes. Your whole library lives in a single portable SQLite file on your machine. No account required. API keys are kept in your OS secure vault (Windows Credential Manager or macOS Keychain), never in a file or a log.
Ogma Library is completely free to download and use; the optional AI advisor uses your own API key. It requires Windows 10 version 1903 or later, or macOS 12 Monterey or later.
Coming soon Windows and macOS builds are in final testing.
Requires Windows 10 version 1903 or later · macOS 12 Monterey or later
Completely free to download and use · the AI advisor uses your own API key