Skip to main content
Luria Voice extensions are distributed as Quarto extensions hosted on GitHub under the brainworkup organization. You install them with the quarto use template command, which scaffolds a complete project, or with quarto add to drop an extension into an existing project without overwriting your files.

System requirements

Make sure these are in place before installing any extension:
Typst is bundled with Quarto. You do not need a separate Typst installation unless you are editing the .typ template files directly.

Scaffold a new project

Use quarto use template to create a new report project from scratch. Quarto downloads the extension, copies the starter template.qmd, and sets up the full directory structure in one step.
Quarto will prompt you to confirm that you trust the template source and ask for a directory name for the new project. After scaffolding, your project directory looks like this:
template.qmd is your starting point. Open it, fill in patient metadata in the YAML front matter, and begin writing your report. See the Quickstart for a full walkthrough.

Add to an existing project

If you already have a Quarto project and want to add a Luria Voice format without scaffolding a new template.qmd, use quarto add. This installs only the extension files under _extensions/ and leaves your existing files untouched.
After adding the extension, reference the format in your document’s front matter:
Then render with:

Update extensions

To pull in the latest changes from GitHub, run the update command from inside your project directory:
Updating an extension overwrites the files under _extensions/. If you have made local edits to typst-template.typ or typst-show.typ, back them up before updating or version-control your project with Git.

R dependencies

Luria Voice report templates use R code chunks for score processing and visualization. You need several CRAN packages and the cingulate package from GitHub.

Install CRAN packages

Install cingulate from GitHub

The cingulate package is not on CRAN. Install it using pak (recommended) or remotes:
pak resolves dependencies faster and handles system library conflicts more gracefully than remotes. If you are setting up a fresh R environment, prefer pak.

Optional: Ollama for AI narratives

The neurotyp-luria extension can generate narrative sections using a locally running Ollama model. Ollama keeps all patient data on your machine — no data is sent to external APIs. Install Ollama from ollama.com, then pull a model:
Once Ollama is running locally, cingulate connects to it automatically when you call the narrative generation functions in your .qmd file. See AI-assisted narratives for configuration details.

Verify your installation

After installing an extension and its dependencies, render the scaffolded template to confirm everything is working:
A successful render produces a PDF in your project directory. If the render fails, check the Troubleshooting page for common errors related to missing R packages, font issues, and Typst compilation problems.
Last modified on May 20, 2026