Package: datablinder 0.1.0

datablinder: Synthetic Look-Alike Copies of Sensitive Data Files

Replaces the contents of a sensitive or proprietary data file with synthetic values while keeping the columns, classes, formats and file type identical, so that analysis code written against the anonymised copy runs unchanged on the real data. Intended for sharing the structure of a dataset with a collaborator or with an AI assistant (a large language model or LLM chatbot) without sharing its contents. Reads and writes delimited text, 'Excel', 'SPSS', 'Stata', 'RDS' and 'Parquet' files, and includes a local 'Shiny' app and a command line script. Reduces disclosure risk; it offers no formal privacy guarantee.

Authors:Alex Gallinat [aut, cre, cph], Zobaer Ahmed [ctb, cph]

datablinder_0.1.0.tar.gz
datablinder_0.1.0.zip(r-4.7-any)datablinder_0.1.0.zip(r-4.6-any)datablinder_0.1.0.zip(r-4.5-any)
datablinder_0.1.0.tgz(r-4.6-any)datablinder_0.1.0.tgz(r-4.5-any)
datablinder_0.1.0.tar.gz(r-4.7-any)datablinder_0.1.0.tar.gz(r-4.6-any)
datablinder_0.1.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
datablinder/json (API)

# Install 'datablinder' in R:
install.packages('datablinder', repos = c('https://agallinat.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/agallinat/datablinder/issues

Pkgdown/docs site:https://agallinat.github.io

On CRAN:

Conda:

aiai-governanceai-safetyanonymizationchatgptdata-privacydata-privacy-compliancedisclosure-controlllmmock-datamock-data-generatorprivacy-protectionprivacy-toolsshinysynthetic-datasynthetic-data-generationsynthetic-dataset-generation

3.48 score 4 exports 54 dependencies

Last updated from:995aacc9d0. Checks:9 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-develOK126
source / vignettesOK250
linux-releaseOK141
macos-releaseOK96
macos-oldrelOK105
windows-develOK109
windows-releaseOK92
windows-oldrelOK98
wasm-releaseOK106

Exports:blind_datablind_fileinstall_clirun_app

Dependencies:base64encbitbit64bslibcachemcellrangerclicliprcommonmarkcpp11crayondata.tabledigestfastmapfontawesomeforcatsfsgluehavenhmshtmltoolshttpuvjquerylibjsonlitelaterlifecyclemagrittrmemoisemimeotelpillarpkgconfigprettyunitsprogresspromisesR6rappdirsRcppreadrreadxlrematchrlangsassshinysourcetoolstibbletidyselecttzdbutf8vctrsvroomwithrwritexlxtable

Other packages for fake and anonymised data
The question that separates them | How the packages line up | FakeDataR, which came first | The rule of thumb
Rendered fromalternatives.Rmdusingknitr::rmarkdown

Last update: 2026-09-29
Started: 2026-09-27

Sharing sensitive data with an AI assistant
Step 1: make the blinded copy | Step 2: read the summary before you share anything | Step 3: give the assistant the file and the summary | Step 4: run what comes back on the real file | Four options that come up in this workflow | Doing it without writing R | Before you send it
Rendered fromsharing-data-with-ai.Rmdusingknitr::rmarkdown

Last update: 2026-09-29
Started: 2026-09-27

What datablinder does not protect you from
It is not a formal privacy guarantee | The structure stays visible, on purpose | The blinded data is for writing code, not for analysis | Some code will not survive the trip | Detection can be wrong, and the summary is where you see it | Exact numeric coincidences are reported, not forbidden | keep_real is the one place the promise is switched off | What the package never does | Not in this version
Rendered fromlimits.Rmdusingknitr::rmarkdown

Last update: 2026-09-29
Started: 2026-09-27

Readme and manuals

Help Manual

Help pageTopics
Blind a data frameblind_data
Blind a data fileblind_file
The summary of a blinded tableblind_summary format.blind_summary print.blind_summary
Install the 'datablinder' commandinstall_cli
Blind a data file in a browserrun_app