quarto-scaffold

Quarto Project Scaffolding

Set up a new Quarto project following RRC team conventions. Ask the user the questions below, then generate the files.

Step 1: Gather Requirements

Ask the user (skip anything they've already answered):

1. Project name — Used for the folder name (snake_case or kebab-case).
2. Output format — HTML report, Word document (.docx), or both?
3. Is this parameterized? — Will the same report be run for multiple sites/parks/datasets? If yes, ask what the key parameters are (e.g., park code, survey year, data file path).
4. Network sync? — Should rendered output be copied to a shared drive (Egnyte/Z: drive)? If yes, include post-render hook setup.

Step 2: Create the Project Structure

Standard folder layout

project-name/
├── _quarto.yml
├── project-name.Rproj
├── .gitignore
├── report.qmd               # or a descriptive name based on the project
├── output/                   # rendered output (gitignored)
└── scripts/                  # helper R scripts

Data typically lives on a network drive (Z: / Egnyte), not in the repo. Data paths are passed via params in the .qmd YAML. The repo contains only scripts and rendered output.

If the project is parameterized with site-specific runs, use the numbered-prefix pattern:

project-name/
├── _quarto.yml
├── project-name.Rproj
├── .gitignore
├── 00_template_scripts/
│   └── report_template.qmd  # master template
├── 01_site_scripts/
│   ├── 01_SITE1/
│   │   └── report.qmd       # site-specific overrides
│   └── 02_SITE2/
│       └── report.qmd
├── output/
└── scripts/

_quarto.yml

For a standard project:

project:
  type: default

If network sync is needed, add the post-render hook:

project:
  type: default
  post-render:
    - "Rscript post_render.R"

Then create post_render.R following the pattern from the quarto-render-sync skill (if available), or note that the user should set it up with /rrc-R-tools:quarto-render-sync.

.Rproj file

Version: 1.0

RestoreWorkspace: No
SaveWorkspace: No
AlwaysSaveHistory: Default

EnableCodeIndexing: Yes
UseSpacesForTab: Yes
NumSpacesForTab: 2
Encoding: UTF-8

AutoAppendNewline: Yes
StripTrailingWhitespace: Yes

.gitignore

# R / RStudio
.Rproj.user/
.Rhistory
.RData
.Ruserdata

# Quarto
/.quarto/
**/*.quarto_ipynb
*_files/

# Rendered output
output/
*.html
*.docx

Step 3: Create the Template .qmd

HTML format

---
title: "Report Title"
date: today
format:
  html:
    theme: cosmo
    toc: true
    toc-depth: 3
    embed-resources: true
editor: visual
execute:
  echo: false
  warning: false
  message: false
---

Word (.docx) format

---
title: "Report Title"
format:
  docx:
    toc: false
    number-sections: true
    fig-width: 6.5
    fig-height: 4
    fig-dpi: 300
editor: visual
execute:
  echo: false
  warning: false
  message: false
---

If the user has a custom reference doc (custom-reference.docx), add reference-doc: custom-reference.docx under the docx: key.

Parameterized version

Add a params block with sensible defaults. Common parameters for RRC projects:

params:
  park: "SITE"
  year_survey: 2026
  data_path: ""
  output_path: ""

Use params in the title with inline R:

title: "`r paste(params$park, params$year_survey)`"
subtitle: "`r paste0('Project Name - ', params$park)`"
date: "`r format(Sys.Date(), '%B %d, %Y')`"

Setup chunk

Every .qmd should start with a setup chunk:

```{r}
#| label: setup
#| include: false

library(tidyverse)
library(haven)
library(scales)

knitr::opts_chunk$set(
  echo = FALSE,
  include = TRUE,
  message = FALSE,
  warning = FALSE
)
```

Add additional libraries based on the project needs:

• Tables: gt for HTML, flextable for Word
• Excel I/O: readxl, writexl
• Spatial/maps: sf, leaflet
• Interactive plots: plotly
• RRC utilities: RrcUtils

Step 4: Confirm with the User

After generating all files, summarize what was created and remind the user:

• Open the .Rproj file in RStudio/Positron to start working
• Update the params block with real values before rendering
• Render with Ctrl+Shift+K or quarto render from the terminal
• If post-render sync was included, set the output_path parameter to the network destination