Re: NSE Documentation System

[Fyodor <fyodor () insecure org>]

On Sun, May 11, 2008 at 09:48:46PM -0700, doug () hcsw org wrote:
> Good to hear this project is being tackled. If you're looking for ideas,
> you might want to take a quick look at the nuff system. Nuff is a platform
> for experimenting with low level network protocols and stands for
> "Network Universal Frame Forge":

Yes, I think the Nuff system is similar to what we want.  I notice
that for each nuff command, you have a description, examples
(sometimes), and required/optional arguments.  The vast majority of
NSE scripts will only have optional arguments (if they have any at
all), but it might be worth designing the system to support required
arguments.  Or maybe it is best to just design the system such that we
can add a required arguments feature if it ever becomes necessary.

Our system should definitely support examples, either as part of the
description or its own section.  I was just thinking how useful that
would be this morning when I had to create some examples for nmap-dev
posts (http://seclists.org/nmap-dev/2008/q2/0293.html and
http://seclists.org/nmap-dev/2008/q2/0295.html ).

> Nuff can parse a program and determine if it does anything that requires

This is key for Nmap too.  Nmap (and/or the scripts themselves) should
be able to programatically obtain at least some of these fields.  For
example, in XML output we might want to include an index of the
descriptions for all of the scripts which produced any output during
the scan.

Cheers,
-F

_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://SecLists.Org