Integrate EnvKey with any language, either in development or on a server, by making your configuration available through the shell as environment variables.
Now that EnvKey v2 has been released, you can find version 2 of envkey-source in a subdirectory of the EnvKey v2 monorepo. Using v2 requires an EnvKey v2 organization (it won't work with ENVKEYs generated in a v1 org).
Here's a guide on migrating from v1 to v2.
envkey-source compiles into a simple static binary with no dependencies, which makes installation a simple matter of fetching the right binary for your platform and putting it in your PATH
. An install.sh
script is available to simplify this, as well as a homebrew tap.
Install via bash:
curl -s https://raw.githubusercontent.com/envkey/envkey-source/master/install.sh | bash
Note: the install.sh script writes, then deletes a couple temporary files to the current directory during installation, so make sure you have write permissions for whatever directory you run this command in. In locked down environments, you may want to run it in $HOME
to be safe.
Another Note: the install.sh script downloads the appropriate envkey-source binary from Github Releases by default, but Github has a fairly low rate limit for unauthenticated requests. For added redundancy, it will fail over to envkey-source-releases.s3.amazonaws.com (an S3 bucket controlled by EnvKey) and download the binary from there if the request to Github Releases fails.
Install manually:
Find the release for your platform and architecture, and stick the appropriate binary somewhere in your PATH
(or wherever you like really).
First, generate an ENVKEY
in the EnvKey App.
Then with a .env
file in the current directory that includes ENVKEY=...
(in development) / an ENVKEY
environment variable set (on a server):
eval $(envkey-source)
Now you can access your app's environment variables in this shell, or in any process (in any language) launched from this shell.
You can also pass an ENVKEY
directly. This isn't recommended for a real workflow, but can be useful for trying things out.
eval $(envkey-source ENVKEY)
If your EnvKey config includes multi-line values, you need to load it with slightly different syntax to preserve formatting. Instead of:
eval $(envkey-source)
echo $SOME_VAR
Use this (note the additional double quotes):
eval "$(envkey-source)"
echo "$SOME_VAR"
--cache cache encrypted config as a local backup (default is true when .env file exists, false otherwise)
--no-cache do NOT cache encrypted config as a local backup even when .env file exists
--cache-dir string cache directory (default is $HOME/.envkey/cache)
--dot-env-compatible change output to .env format
--env-file string ENVKEY-containing env file name (default ".env")
--pam-compatible change output format to be compatible with /etc/environment on Linux
-f, --force overwrite existing environment variables and/or other entries in .env file
-h, --help help for envkey-source
-v, --version prints the version
--verbose print verbose output (default is false)
--timeout float timeout in seconds for http requests (default 10)
--retries uint8 number of times to retry requests on failure (default 3)
--retryBackoff float retry backoff factor: {retryBackoff} * (2 ^ {retries - 1}) (default 1)
If you get an error, envkey-source will echo the error string to stdout and return false instead of setting environment variables. For example:
$ eval $(envkey-source notvalidenvkey) && ./env-dependent-script.sh
error: ENVKEY invalid
Whenever you use eval
, you need to worry about shell injection. We did the worrying for you--envkey-source wraps all EnvKey variables in single quotes and safely escapes any single quotes the variables might contain. This removes any potential for shell injection.
By default, envkey-source will not overwrite existing environment variables or additional variables set in a .env
file. This can be convenient for customizing environments that otherwise share the same configuration. But if you do want EnvKey vars to take precedence, use the --force
/ -f
flag. You can also use sub-environments in the EnvKey App for this purpose.
envkey-source caches your encrypted config in development so that you can still use it while offline. Your config will still be available (though possibly not up-to-date) the next time you lose your internet connection. If you do have a connection available, envkey-source will always load the latest config.
By default, caching is enabled when a .env
file is present in the directory, and disabled otherwise. You can also enable it with the --cache
flag or disable it with the --no-cache
flag.
Assume you have GITHUB_TOKEN
set to cf4b78a2b8356059f340a7df735d0f63
for the development
environment in the EnvKey App. You generate a local development ENVKEY
.
In your project's .env
file (ignored from source control):
# .env
ENVKEY=GsL8zC74DWchdpvssa9z-nk7humd7hJmAqNoA
Run envkey-source:
$ eval $(envkey-source)
Now GITHUB_TOKEN
is available in the shell:
$ echo $GITHUB_TOKEN
cf4b78a2b8356059f340a7df735d0f63
Or in any process you launch from this shell:
$ python
>>> import os
>>> os.environ["GITHUB_TOKEN"]
'cf4b78a2b8356059f340a7df735d0f63'
You can do exactly the same on a server, except instead of putting your ENVKEY
in a .env
file, you'll set it as an environment variable (in whatever way you set environment variables for your host/server management platform).
So you set an environment variable on your server:
ENVKEY=HSyahYDL2jBpyMnkV6gF-2rBFUNAHcQSJTiLA
Then you run envkey-source as part of your server start and restart commands, whatever those may be.
$ eval $(envkey-source) && server-start
$ eval $(envkey-source) && server-restart
If you're using envkey-source on a CI server, the process is much the same. Set the ENVKEY
environment variable in your CI interface, then run eval $(envkey-source)
before running tests.
Here's a simple example using Python:
FROM python:3
# install envkey-source
RUN curl -s https://raw.githubusercontent.com/envkey/envkey-source/master/install.sh | bash
RUN mkdir /code
WORKDIR /code
ADD . /code/
# set EnvKey environment variables before running the process
CMD eval $(envkey-source) && python3 example.py
To supply the ENVKEY
in development with docker-compose, you can add it to a .env
file, then use the env_file
key in docker-compose.yml
.
services:
example:
build: .
env_file: .env
On a server, you just need to pass the ENVKEY environment variable through to your docker container. Where to set this depends on your host, but it shouldn't be difficult.
And now you can access EnvKey variables the same way you'd read normal environment variables.
# example.py
import os
print(os.environ["GITHUB_TOKEN"])
Note that if you run envkey-source inside a script, your environment variables will only be visible to commands run within that script unless you run the script with source
, in which case they will be set in the current shell.
envkey-source works well with direnv. Just add the following to your .envrc
file:
export ENVKEY=HSyahYDL2jBpyMnkV6gF-2rBFUNAHcQSJTiLA
if has envkey-source; then
eval $(envkey-source --cache)
fi
and rerun direnv allow
.
On a stripped down OS like Alpine Linux, you may get an x509: certificate signed by unknown authority
error when envkey-source
attempts to load your config. envkey-fetch (which envkey-source
wraps) tries to handle this by including its own set of trusted CAs via gocertifi, but if you're getting this error anyway, you can fix it by ensuring that the ca-certificates
dependency is installed. On Alpine you'll want to run:
apk add --no-cache ca-certificates
envkey-fetch - lower level command line tool that simply accepts an ENVKEY
and spits out decrypted config as json. Handles core fetching, decryption, verification, web of trust, redundancy, and caching logic. Does most of the work behind the scenes for this library.
envkey-ruby - EnvKey Client Library for Ruby and Rails.
envkey-node - EnvKey Client Library for Node.js.
envkeygo - EnvKey Client Library for Go.
For more on EnvKey in general:
Read the docs.
Read the integration quickstart.
Read the security and cryptography overview.
Post an issue or email us: [email protected].