Ok, that looks promising. So, I imagine the workflow like this:
1) An Alert occurs (say disk space at 95%)
2) Nagios runs a command i have set up in command management to fire the notification to my application
3) The user responds with "Ack"
4) My application adds a line to the command file, let's say it is a "ACKNOWLEDGE_HOST_PROBLEM" command.
5) Nagios wakes up and reads this file and performs the actions specified in the ACKNOWLEDGE_HOST_PROBLEM details.
Sound about right?