> ## Documentation Index
> Fetch the complete documentation index at: https://docs.brainworkup.org/llms.txt
> Use this file to discover all available pages before exploring further.

# Luria Voice system requirements and dependencies

> Full list of required and optional software for Luria Voice, including versions, download links, and R package install commands.

Luria Voice depends on a small set of external tools and R packages. The table below lists every dependency, its minimum version, whether it is required or optional, and where to get it. Detailed install instructions follow.

## Requirements at a glance

| Dependency  | Minimum version      | Required | Notes                                            |
| ----------- | -------------------- | -------- | ------------------------------------------------ |
| Quarto      | 1.4.0                | Yes      | Bundles Typst — no separate Typst install needed |
| Typst       | Bundled with Quarto  | Yes      | Do not install separately unless debugging       |
| R           | 4.2.0                | Yes      | Earlier versions may work but are untested       |
| dplyr       | any CRAN             | Yes      | Data manipulation                                |
| readr       | any CRAN             | Yes      | CSV / data file reading                          |
| here        | any CRAN             | Yes      | Portable file paths                              |
| yaml        | any CRAN             | Yes      | YAML front-matter parsing                        |
| ggplot2     | any CRAN             | Yes      | Figure generation                                |
| systemfonts | any CRAN             | Yes      | Font detection for Typst                         |
| cingulate   | GitHub (brainworkup) | Yes      | Report pipeline utilities                        |
| Ollama      | any                  | No       | Required only for AI narrative generation        |

## Quarto

Quarto ≥ 1.4.0 is required. Quarto 1.4 introduced the Typst output format that Luria Voice depends on. Download the installer for your operating system from [quarto.org](https://quarto.org).

After installing, confirm your version:

```bash theme={null}
quarto --version
```

The output should be `1.4.0` or higher.

<Note>
  Quarto ≥ 1.4.0 ships with Typst embedded. You do not need to install Typst
  separately. If you have a standalone Typst installation, Quarto uses its own
  bundled copy and the two will not conflict.
</Note>

## R

R 4.2 or later is recommended. Download from [cran.r-project.org](https://cran.r-project.org).

Confirm your version in R:

```r theme={null}
R.version$major
R.version$minor
```

## R packages

Install all required CRAN packages in a single command:

```r theme={null}
install.packages(c(
  "dplyr",
  "readr",
  "here",
  "yaml",
  "ggplot2",
  "systemfonts"
))
```

### cingulate (GitHub)

The `cingulate` package provides the scoring, norming, and domain-assembly pipeline that populates report sections. It is not on CRAN and must be installed from GitHub.

<Steps>
  <Step title="Install the remotes package">
    If you do not already have `remotes`, install it from CRAN:

    ```r theme={null}
    install.packages("remotes")
    ```
  </Step>

  <Step title="Install cingulate from GitHub">
    ```r theme={null}
    remotes::install_github("brainworkup/cingulate")
    ```
  </Step>

  <Step title="Verify the installation">
    ```r theme={null}
    library(cingulate)
    ```

    If this command runs without error, `cingulate` is installed correctly.
  </Step>
</Steps>

<Warning>
  `remotes::install_github` requires an internet connection and access to
  GitHub. If you are behind a corporate firewall, contact your IT department to
  allow outbound connections to `api.github.com` and `codeload.github.com`.
</Warning>

## Ollama (optional)

Ollama enables AI-generated narrative summaries within the report. It is entirely optional — all scoring, tables, and figures render without it.

If you want to use AI narratives:

<Steps>
  <Step title="Download and install Ollama">
    Download Ollama from [ollama.com](https://ollama.com) and follow the installer for your platform.
  </Step>

  <Step title="Pull a model">
    ```bash theme={null}
    ollama pull llama3
    ```
  </Step>

  <Step title="Start the Ollama server before rendering">
    ```bash theme={null}
    ollama serve
    ```

    Ollama must be running on `localhost:11434` when you call `quarto render`. If it is not running, the narrative generation step will time out and log an error.
  </Step>
</Steps>

## Installing everything at once

The following block installs all R dependencies in a single session. Run it once after a fresh R installation.

<CodeGroup>
  ```r Install all R dependencies theme={null}
  install.packages(c(
    "dplyr",
    "readr",
    "here",
    "yaml",
    "ggplot2",
    "systemfonts",
    "remotes"
  ))

  remotes::install_github("brainworkup/cingulate")
  ```

  ```bash Verify Quarto version theme={null}
  quarto --version
  # Expected: 1.4.x or higher
  ```
</CodeGroup>