When authorizing SSL keys, you typically concatenate it to the authorized_keys file. Like this:

cat publickey.pub >> ~/.ssh/authorized_keys

But this simple method doesn't work when you're trying to load keys automatically .The keys might already exist, and if you keep concatenating the file, authorized_keys will get miles long with duplicates. So, here's a basic bash script to authorize a key if it doesn't already exist in the index:

#!/bin/bash
keypub="/root/github.pub"
keyprivate="/root/github"
while read line; do 
	keydata=$line
done < $keypub
keydata="${keydata#"${keydata%%[![:space:]]*}"}"
keydata="${keydata%"${keydata##*[![:space:]]}"}"
keyfound=`fgrep "${keydata}" ~/.ssh/authorized_keys`
if [[ -n "${keyfound}" ]]; then
	echo "${keydata}"
	echo "SSH key already loaded into ~/.ssh/authorized_keys"
	exit 1
else
	keypubname=`basename $keypub`
	keyprivatename=`basename $keyprivate`
	sudo cp -f "$keyprivate" "/root/.ssh/${keyprivatename}"
	sudo cp -f "$keypub" "/root/.ssh/${keypubname}"
	sudo cat "/root/.ssh/${keypubname}" >> "/root/.ssh/authorized_keys"
	echo "${keydata}"
	echo "Key successfully loaded into ~/.ssh/authorized_keys"
	exit 0
fi

Update these values to point to the public/private keys:

keypub="/root/github.pub"
keyprivate="/root/github"

Bash is beautiful for automation, isn't it?