Cloudwork is a bare-bones automation tool for quickly spawning machines in the cloud and distributing workloads across them.
Warning
Screen recording is outdated.
first.mp4
Requirements:
sshrsync
go install github.com/3rd/cloudwork@latestFirst, you'll need to create a cloudwork.yml configuration file.
The configuration file is a YAML file that looks like this:
workers:
- host: worker1
- host: worker2
- host: worker3
scripts:
setup: |
# This runs when you do `cloudwork setup` if the worker wasn't already setup or if the script changed.
apt update -y && apt upgrade -y
apt install -y apt-transport-https ca-certificates curl software-properties-common gcc clang make build-essential libssl-dev libffi-dev libpcap-dev
apt install -y neovim
curl -s https://get.docker.com/ | sh
usermod -aG docker $USER
mkdir /app
upload Dockerfile /app
default: |
# This runs on each worker when you do `cloudwork run`.
upload-input /tmp/worker/input/
docker build -t work /app
docker run -v /tmp/worker/input:/input -v /tmp/worker/output:/output -it work
download-output /tmp/worker/output/Use cloudwork bootstrap to create the input/output directory structure for the configured workers.
After running this command, you will have a "workers" directory with the following structure:
|-- workers/
| |-- worker1/
| | |-- input/*
| | |-- output/*
| |-- worker2/
| | |-- input/*
| | |-- output/*
| ...
|-- cloudwork.yml
In the example configuration, setup and run are scripts that will run on workers.
There are a few special commands that you can use in scripts:
upload-input <remote path>- Uploads./workers/<worker/input/*to the worker at<remote path>.download-output <remote path>- Downloads<remote path>to./workers/<worker/output/.upload <local path> <remote path>- Uploads localhost:<local path>to remote:<remote path>.download <remote path> <local path>- Downloads remote:<remote path>to localhost:<local path>.
Warning
Upload and download operations are extracted from the script and don't run when you'd expect. Uploads are done before the script is run, and downloads are done after the script is run.
Tip
The paths used in the upload/download commens are pass as-they-are to rsync, remember to add a trailing slash if you want to upload/download the contents of a directory.
cloudwork bootstrap- Creates the input/output directory structure for the configured workers.cloudwork run [script-name]- Runs the specified script on all workers. If no script name is provided, it runs the "default" script.cloudwork run ./script.sh- Runs a script (file) on all workers.cloudwork exec "command"- Executes a command on all workers.couldwork -host <host> <command>- idem, but only for the specified<host>.
workers:
- host: worker1
- host: worker2
- host: worker3
scripts:
setup: |
apt update -y && apt upgrade -y
apt install -y apt-transport-https ca-certificates curl software-properties-common gcc clang make build-essential libssl-dev libffi-dev libpcap-dev
apt install -y golang-go
mkdir -p "$HOME/go"
export GOPATH="$HOME/go"
export PATH="$GOPATH/bin:$PATH"
grep -q "export GOPATH" ~/.profile || {
echo "export GOPATH=$HOME/go" >>~/.profile
echo "export PATH=\$GOPATH/bin:\$PATH" >>~/.profile
}
git clone https://github.com/blechschmidt/massdns.git
cd massdns
make
make install
go install github.com/d3mondev/puredns/v2@latest
mkdir -p /worker
upload sample-wordlist.txt /worker
upload sample-resolvers.txt /worker
default: |
puredns bruteforce \
/worker/sample-wordlist.txt \
-t 100 -rate-limit 500 \
-r /worker/sample-resolvers.txt \
-d /tmp/worker/input/domains.txt \
-w /tmp/worker/output/results.txt \
--write-wildcards /tmp/worker/output/wildcards.txt