Local-first document tools for reading, cleaning, collecting, converting, and organizing PDFs, Markdown files, mail attachments, invoices, and research libraries.
doc-bricks builds practical desktop applications for document-heavy workflows. The projects focus on local document libraries, safe mail processing, OCR, PDF export, literature management, and structured handoff formats that remain useful for humans and LLM-assisted maintenance.
Public index status: verified on 2026-06-26 against the live GitHub organization. The profile lists 10 public repositories total: .github plus 9 product/tool repositories. Private or internal repositories are intentionally not listed.
| Need | Start with | Why |
|---|---|---|
| Read, organize, and export a local document library | DokuReader | Topic-based document manager with previews, reading state, PDF bundling, and metadata-only JSON export |
| Work with Markdown notes and clean reading sessions | CleanMarkdown | Local Markdown viewer/editor with reading mode, raw editing, PDF export, and PWA companion |
| Manage literature, PDFs, BibTeX, and research notes | LitZentrum | Literature manager for academic reading, bibliography workflows, and PDF-centered research |
| Download and archive mail attachments locally | UniversalDocsGrabber | IMAP/Gmail attachment downloader with OCR, PDF conversion, dedupe, and review workflows |
| Clean large IMAP or Gmail mailboxes safely | UniversalMailCleaner | Mailbox cleaner with safe trash mode, labels, scheduler, and large-item cleanup |
| Collect invoices and receipts from mail | UniversalInvoiceMail | Invoice extractor with Gmail/IMAP input, OCR, PDF conversion, JSON export, and DATEV-oriented workflows |
| Keep local notes for LLM agents | llm-note | Local-first SQLite notes, plain-text notebooks, six locales, and a standalone agent skill extracted from BACH |
This index covers every public doc-bricks repository visible on GitHub as of 2026-06-26. Private or internal repositories are intentionally not listed on the public organization profile.
| Repository | Role | Discovery notes |
|---|---|---|
| .github | Organization profile and shared community-health files | Start page, issue templates, pull request template, security policy, and llms.txt |
| DokuReader | Document library | Local-first document manager, reading state, topic organization, preview workflow, PDF bundling |
| CleanMarkdown | Markdown reading and editing | Markdown viewer/editor, reading mode, PDF export, session handoff, PWA companion |
| LitZentrum | Literature management | PDF library, BibTeX workflows, academic reading, research notes, JSON export |
| UniversalDocsGrabber | Mail attachment intake | IMAP/Gmail attachment downloader, OCR, PDF conversion, dedupe, document archive review |
| UniversalMailCleaner | Mailbox cleanup | Gmail and IMAP cleaner, safe trash mode, labels, scheduler, large-item cleanup |
| UniversalInvoiceMail | Invoice and receipt intake | Invoice extractor, OCR, PDF conversion, JSON export, DATEV-oriented archive workflows |
| MailProcessor | Mail tool launcher | System tray entry point for Universal Mail Cleaner, UniversalDocsGrabber, and UniversalInvoiceMail |
| MediaBrain | Media and document hub | Local media library, browser activity context, research material organization |
| llm-note | Agent notes and notebooks | Local-first SQLite note log, plain-text notebooks, six locales, CLI/Python API, and standalone agent skill |
| App | Description |
|---|---|
| DokuReader | Local-first document library with topic organization, previews, read status, PDF bundling, and metadata export |
| CleanMarkdown | Markdown viewer/editor for clean reading, editing, PDF export, and session handoff |
| LitZentrum | Literature management suite with PDF, BibTeX, JSON export, and research-oriented organization |
| MediaBrain | Local media and document hub for files, browser activity, and research material |
| App | Description |
|---|---|
| MailProcessor | System tray launcher for Universal Mail Cleaner, UniversalDocsGrabber, and UniversalInvoiceMail |
| UniversalMailCleaner | Local Gmail and IMAP mailbox cleanup with safe modes and scheduler support |
| UniversalDocsGrabber | Attachment downloader and document archive builder for mail-based document intake |
| UniversalInvoiceMail | Invoice and receipt collection pipeline for local archives and accounting-oriented exports |
| App | Description |
|---|---|
| UniversalDocsGrabber | Converts downloaded mail documents into reviewable PDF/OCR archive material |
| UniversalInvoiceMail | Converts invoice and receipt mail into PDF/OCR archive material and accounting-oriented exports |
| CleanMarkdown | Exports Markdown reading sessions to PDF and companion web formats |
| App | Description |
|---|---|
| llm-note | Local-first notes and notebook inboxes for LLM agents, extracted from BACH Notizblock/Denkarium patterns |
- Local first: documents, mail exports, indexes, and reading state stay on the user's machine by default.
- Privacy-conscious: tools avoid cloud processing unless an external service such as Gmail is explicitly configured by the user.
- Document-practical: each project targets repeated real workflows: reading, previewing, cleaning, exporting, archiving, and handoff.
- Structured exports: JSON, BibTeX, PDF, and companion formats are documented so data can move between desktop tools and later automation.
- Readable maintenance: READMEs, tests, and
llms.txtfiles are kept useful for both human maintainers and LLM-based assistants.
Useful search phrases for the public doc-bricks profile include local-first document tools, Python document management, PySide6 PDF OCR, Gmail IMAP attachment downloader, Markdown PDF export, literature manager GitHub, invoice mail OCR DATEV export, and local-first LLM agent notes.
For crawlers and LLM tools, see llms.txt. It lists the canonical repositories, project roles, and preferred search phrases for the doc-bricks organization.
doc-bricks is the document-work branch of the brick suite:
open-bricks | file-bricks | dev-bricks
Part of the ellmos-ai ecosystem.
| Project | Description |
|---|---|
| WikiStub-Seed | Multilingual knowledge-stub dataset (630 terms, 12 domains, DE/EN + es/ja/ru/zh) with a web publisher — deployable as a self-contained wiki module |