Lixity

Lixity is a local workbench for authors. Explore language patterns and keep sources, claims and notes beside your manuscript. Try an invented example in your browser before installing the app for your own writing.

Open the example manuscript · Take the short tour · Install Lixity

The read-only example uses six invented English chapters and credited research images. It needs no account or upload and cannot import, edit or save. More views and image credits.

Install and start

Install uv, then run on Windows, macOS or Linux:

uv tool install --python 3.12 --managed-python lixity
uv tool update-shell

Open a new terminal:

lixity --version
lixity service start --open

Choose Open Project or Analyze a manuscript, then confirm the language. English is the default; automatic detection is explicit. Built-in analysis works offline on your computer and does not grade or rewrite prose. Optional integrations have separate data boundaries.

Your first project explains the workflow. Installation and updates covers launchers, pinned wheels, service management and removal. After upgrading, run lixity service restart --open and reload the browser; installing a package does not update an already running service.

What you can do

• Manuscript & Analysis. Open paragraph values and work markers, or compare chapters against the manuscript's own patterns. Analysis can be turned on and off.
• Research & Dossiers. Link retained sources, authored claims and decisions to exact manuscript quotations. Keep a dossier alongside a passage; Zotero remains your collection tool. Retention is explicit, and archive integrity does not establish factual truth.
• Project & exports. Manage project metadata and create editorial, reading, submission or analysis PDFs, terminal reports and versioned JSON. NDA generation creates a draft without recording consent or signatures; dossier Markdown edits require explicit preview and apply. See export and project limits.

The browser app, CLI and Python API share one analysis pipeline; lixity --help and lixity about --json describe installed capabilities.

Turn Analysis on beside the text, choose a metric and inspect its counted passages through hover, touch or the paragraph menu. The same literal evidence is available through lixity profile manuscript.md --layer asl --evidence. A passage's Saved research opens its saved claim and dossier versions beside the manuscript, with a return to the passage. See the reader and CLI guide.

CLI workflows can reuse the project's configured manuscript; default manuscript guidance explains explicit project selection and settings scope.

English and German have the broadest analysis resources. French, Spanish, Italian, Portuguese and Dutch offer localized interfaces with narrower heuristic coverage. Language support and interpretation explain coverage, short-sample limits and unavailable measurements.

Documentation map

In the repository, src/lixity/ contains the shared engine, CLI and UI; tests/ verifies behavior; docs/ holds maintained product documentation; and samples/ contains synthetic or public-domain examples. Book-specific adapters belong in their own projects.

Release, citation and license

The current release is v0.7.3, in the pre-1.0 development line. Historical v3.0.1 and earlier tags remain available. Product and schema versions are independent; Release notes cover archive compatibility and upgrades. See the changelog for subsequent work. The public demo follows current main and can show changes beyond the release.

Cite Lixity on Zenodo, with metadata in CITATION.cff. The software paper draft describes its research context.

AGPL-3.0-or-later. Use Lixity for private or commercial writing, research and teaching, including self-publishing your book. A separate written commercial license is available for integrations that cannot meet the AGPL. Licensing guide · Full terms · Legal notice and contact.