A Simple Client to the 'AWS' Key Management Service

Encrypt plain text and 'decrypt' cipher text using encryption keys hosted at Amazon Web Services ('AWS') Key Management Service ('KMS'), on which see < https://aws.amazon.com/kms> for more information.

This an R client to interact with the AWS Key Management Service, including wrapper functions around the KMS Java client to encrypt plain text and decrypt cipher using Customer Master Keys stored in KMS.


CRAN version

The package is hosted on CRAN, so installation is as easy as:


But you can similarly easily install the most recent development version of the R package as well:


This R package relies on the jar files bundled with the AWR package.

What is it good for?

Currently, only three basic, but very important features are supported:

  • you can encrypt up to 4 KB of arbitrary data such as an RSA key, a database password, or other sensitive customer information and Base64-encode it to be stored somewhere:
> kms_encrypt('alias/mykey', 'foobar')
[1] "Base-64 encoded ciphertext"
  • decrypt such Base-64 encoded ciphertext back to plaintext:
> kms_encrypt('Base-64 encoded ciphertext')
[1] "foobar"
  • generate a data encryption key (see below for a use case):
> kms_generate_data_key('alias/mykey')
[1] "Base-64 encoded, encrypted data encryption key"
[1] "alias/mykey"
[1] "Base-64 encoded data encryption key"

How can I encrypt data larger than 4KB?

Use envelope encryption. In short, you can:

  • generate a new (data) encryption key (eg with kms_generate_data_key) and store it only in memory for the next 2 steps
  • use this new encryption key to encrypt the data locally (eg using the sodium package or the AES function from the digest package)
  • encrypt the encryption key via KMS and store the encrypted (data encryption) key on disk along with the encrypted data
  • clean up the encryption key from memory
  • if you want to decrypt the data, decrypt the encrypted (data encryption) key via KMS, than decrypt the data with the decrypted (data encryption) key stored in memory

A simple implementation:

## let's say we want to encrypt the mtcars dataset stored in JSON
data <- toJSON(mtcars)
## generate a 256-bit data encryption key (that's supported by digest::AES)
key <- kms_generate_data_key('alias/mykey', byte = 32L)
## convert the JSON to raw so that we can use that with digest::AES
raw <- charToRaw(data)
## the text length must be a multiple of 16 bytes
## https://github.com/sdoyen/r_password_crypt/blob/master/crypt.R
raw <- c(raw, as.raw(rep(0, 16 - length(raw) %% 16)))
## encrypt the raw object with the new key + digest::AES
## the resulting text and the encrypted key can be stored on disk
aes <- AES(key$text)
## decrypt the above returned ciphertext using the decrypted key
rawToChar(aes$decrypt(base64_dec(...), raw = TRUE))

What if I want to do other cool things with KMS and R?

Writing wrapper functions around the Java SDK is very easy. Please open a ticket on the feature request, or even better, submit a pull request :)

It doesn't work here!

To be able to use this package, you need to have an AWS account and a KMS Encryption Key. If you do not have one already, you can register for free at Amazon and do 20K free requests per month, although keys do cost 1 USD per month.

Once you have an AWS account, make sure your default AWS Credentials are available via the DefaultAWSCredentialsProviderChain . In short, you either provide a default credential profiles file at ~/.aws/credentials, use the AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY environment variables or if using AWR.KMS on AWS, you can also rely on the EC2 instance profile credentials or ECS Task Role as well.


Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.1 by Gergely Daroczi, 4 years ago


Browse source code at https://github.com/cran/AWR.KMS

Authors: Gergely Daroczi <[email protected]>

Documentation:   PDF Manual  

AGPL-3 license

Imports AWR, rJava, jsonlite

See at CRAN