Changes in version 0.2.0                        

New features

  - init_template() now accepts any GitHub repository as "owner/repo" or
    "owner/repo@ref" (branch, tag, or commit), in addition to the
    built-in names. Templates are defined in a data-file registry, and
    the download targets the repository's default branch (HEAD) instead
    of a hardcoded main.

  - peacock_templates() lists the built-in templates. A custom registry
    (DCF or YAML) can be supplied via the registry argument.

  - init_analysis() scaffolds a reproducible analysis project
    (raw/processed data, analysis notebooks, reusable functions, and
    outputs).

  - init_quarto() scaffolds a Quarto website, book, or manuscript.

  - init_python() emits a modern Python project (src layout,
    pyproject.toml with ruff and pytest). peacock stays an R package.

  - Every project scaffold - init_shiny(), init_analysis(),
    init_quarto(), init_python(), and tool_review_template() - now also
    writes AGENTS.md and CLAUDE.md so AI coding assistants are
    productive immediately.

Reliability

  - Added a testthat (edition 3) test suite and a cross-platform R CMD
    check GitHub Actions workflow (Ubuntu, Windows, macOS).

  - Adopted the Air formatter, enforced in CI.

Bug fixes

  - init_template() validates its template argument and fails fast with
    a clear message for unknown names; the download is wrapped with
    error handling and always cleans up its temporary directory.

  - tool_review_template() uses seq_along() (previously errored on an
    empty tool_name) and now populates new and empty src/ scripts.

  - The Shiny Dockerfile CMD uses valid quoting (previously backticks).

  - Confirmation prompts are shown only in interactive sessions, so
    scripted use no longer silently cancels.

Other

  - peacock is now released under the MIT license (previously GPL-3).

  - Added a package hex-sticker logo (in an Indian folk-art style), a
    companion overview graphic, and browser-tab favicons for the
    documentation site.

  - The init_* scaffolding functions now return the created path
    invisibly.

                        Changes in version 0.1.0                        

Initial release of peacock.

Features

  - init_shiny() - Initialize a complete Shiny application structure
    with organized folders for modules, UI components, data, and static
    assets. Includes Dockerfile for deployment.

  - init_template() - Download and set up project templates from GitHub
    repositories. Currently supports "shiny" and "cgds" templates.

  - init_changelog_md() - Create a CHANGELOG.md file with a simple
    format for tracking project progress and changes.

  - tool_review_template() - Set up an organized structure for comparing
    multiple tools or methods, with dedicated folders for each tool's
    data, scripts, and outputs.

Notes

  - All functions include a confirmation prompt by default. Set confirm
    = FALSE to skip prompts in automated workflows.

  - Templates create standard directory structures that can be modified
    after creation to fit specific needs.

  - Package includes built-in error handling and user-friendly messages
    for setup workflows.