Skip to contents

peacock

peacock sets up a ready-to-work project directory in one call: the layout, the config files, and the agent guidance. It scaffolds R projects, and also Quarto and Python ones, so the same habit works across a mixed stack.

One function scaffolds a ready-to-work project: init_shiny('my_app') creates ui.R, server.R, global.R, R/, www/ and an AGENTS.md

Installation

From r-universe:

install.packages("peacock", repos = "https://samuelbharti.r-universe.dev")

Or the development version:

# install.packages("pak")
pak::pak("samuelbharti/peacock/pkg-r")

What can peacock do?

1. Initialize a Shiny app structure

Creates a complete Shiny app with organized folders and starter files.

library(peacock)

# Set up a new Shiny project
init_shiny(path = "my_shiny_app")

This creates:

  • ui.R, server.R, global.R - Your main Shiny files
  • modules/ - For modular Shiny components
  • userInterface/ - UI components organized separately
  • www/ - Static files (CSS, images, JavaScript)
  • data/, R/, dev/ - Standard project folders
  • Dockerfile - Ready for containerization
  • .gitignore and .Renviron - Project configuration

Set confirm = FALSE to skip the confirmation prompt, which is what you want in a script.

2. Create a changelog

Track project changes in a structured markdown file.

init_changelog_md(path = "my_project")

Creates a CHANGELOG.md file with a simple format for documenting your work.

3. Use GitHub templates

Pull down complete project templates from GitHub. Use a built-in name, or any GitHub repository as "owner/repo" (optionally pinned with "owner/repo@ref").

# Built-in templates
init_template("shiny", path = "my_project")
init_template("cgds", path = "research_project")

# Any GitHub repo, optionally pinned to a branch, tag, or commit
init_template("owner/repo", path = "my_project")
init_template("owner/repo@dev", path = "my_project")

# See the built-in templates
peacock_templates()

The CGDS template is part of the UAB CGDS research workflow.

4. Set up tool review projects

Organize projects that compare multiple tools or methods.

tool_review_template(
  tool_name = c("tool1", "tool2", "tool3"),
  tool_url = c("https://tool1.com", "https://tool2.com", "https://tool3.com"),
  path = "tool_comparison"
)

This creates organized folders for data, scripts, outputs, and documentation for each tool.

5. Scaffold a reproducible analysis project

Set up a tidy layout for a data-analysis or research project.

init_analysis(path = "my_study")

Creates data/{raw,processed}, R/, analysis/ (with a starter Quarto notebook), and output/{figures,tables}, plus a README and .gitignore.

6. Scaffold a Quarto project

Create a Quarto website, book, or manuscript.

init_quarto(path = "my_site", type = "website")

type can be "website" (default), "book", or "manuscript"; each gets a _quarto.yml and starter documents.

7. Scaffold a Python project

peacock stays an R package, but it can emit a Python project too.

init_python(path = "my_pkg")

Creates a src-layout package with pyproject.toml (ruff + pytest), a starter module and test, .gitignore, and AGENTS.md.

Every project is AI-ready

peacock’s project scaffolds all drop an AGENTS.md (plus a CLAUDE.md that imports it) describing how to run, build, and work in the project, so AI coding assistants are productive from the first commit.

RStudio Integration

Once installed, peacock adds an RStudio Add-in for quick access:

  • Find “Peacock: Shiny Template” in the Addins menu
  • Or create a new project and select “Peacock: Shiny Template” from the templates

Learn more

License

MIT © Samuel Bharti