R Documentation Scraper

Scrape R package metadata from RDocumentation.org / CRAN using the public crandb API. No authentication or proxies required.

What it does

• Search packages by keyword — finds all CRAN packages matching your query
• Fetch specific packages by exact CRAN package name
• Returns rich metadata: version, title, description, author, maintainer, license, dependencies, imports, and canonical RDocumentation links

Input

Field	Type	Description	Default
mode	string	searchPackages or getByName	searchPackages
searchQuery	string	Keyword to search (mode=searchPackages)	ggplot
packageNames	string[]	Exact package names (mode=getByName)	[]
maxItems	integer	Max records to return (1–500)	50

Example: search packages

{
  "mode": "searchPackages",
  "searchQuery": "ggplot",
  "maxItems": 20
}

Example: get by name

{
  "mode": "getByName",
  "packageNames": ["ggplot2", "dplyr", "tidyr", "data.table"],
  "maxItems": 10
}

Output

Each record contains:

Field	Type	Description
packageName	string	CRAN package name
version	string	Latest version
title	string	Short one-line title
description	string	Full description
author	string	Package authors
maintainer	string	Current maintainer with email
license	string	License string (e.g. MIT, GPL-2)
type	string	Repository type (CRAN, Bioconductor)
depends	string[]	Packages listed in Depends (R excluded)
imports	string[]	Packages listed in Imports
repositoryUrl	string	First URL from URL field
bugReportsUrl	string	Bug tracker URL
sourceUrl	string	https://www.rdocumentation.org/packages/{name}
publishedAt	string	Publication date if available
recordType	string	Always "package"
scrapedAt	string	ISO 8601 timestamp of scrape

Sample output record

{
  "packageName": "ggplot2",
  "version": "4.0.3",
  "title": "Create Elegant Data Visualisations Using the Grammar of Graphics",
  "description": "A system for 'declaratively' creating graphics, based on The Grammar of Graphics...",
  "author": "Hadley Wickham [aut], Winston Chang [aut], Thomas Lin Pedersen [aut, cre]...",
  "maintainer": "Thomas Lin Pedersen <thomas.pedersen@posit.co>",
  "license": "MIT + file LICENSE",
  "type": "CRAN",
  "imports": ["cli", "grDevices", "grid", "gtable", "isoband", "lifecycle", "rlang", "scales"],
  "repositoryUrl": "https://ggplot2.tidyverse.org",
  "bugReportsUrl": "https://github.com/tidyverse/ggplot2/issues",
  "sourceUrl": "https://www.rdocumentation.org/packages/ggplot2",
  "recordType": "package",
  "scrapedAt": "2026-05-30T10:00:00+00:00"
}

FAQs

Q: Does this require an API key or proxy? No. The scraper uses the public crandb.r-pkg.org API, which is free and open.

Q: Does this return ALL versions of a package? No. It returns metadata for the latest published version only.

Q: How does search work? The scraper performs a prefix-range query on the crandb API, returning all packages whose names start with your search term. It also falls back to a substring match across all CRAN package names.

Q: How many packages are on CRAN? As of 2026, CRAN hosts over 20,000 packages.

Q: Can I fetch Bioconductor packages? Currently the scraper covers CRAN only (via crandb). Bioconductor packages may not appear.

Q: What is the rate limit? The crandb API is generous for public usage. The scraper adds small delays between requests to be polite.