Net::Twitter has an OAuth role that you can use:

Take a look at the examples section on that page. The first example
demonstrates a typical pattern for authorizing the client and storing
the access token on disk.

One complication you're going to run into is that people who use your
client will need to generate their own consumer key/secret pairs,
because you can't distribute your secret in the open. Take a look at
the "authorize" command in Marcel and Raffi's twurl client
( for a good example of how to handle


On Mon, May 17, 2010 at 6:34 AM, Neal Rauhauser <> wrote:
>  I've published a handful of 'unix way' scripts written in perl and
> based on Marc Mims' Net::Twitter library. The basic idea was to lower
> the barrier of entry for folks who want some simple thing done - say a
> unix admin who wants a system to be able to tweet/DM. There are a lot
> of things that could be done with this - I've played a bit with
> integrating Nagios alerts using this, with making Request Tracker able
> to send messages via Twitter, etc.
>    These are little things, as they should be, and made to run from
> the command line, which is the right way to handle lots of text data.
> I'm seeing this stuff on Oauth, I'm absolutely not a web developer in
> any sense of the word, and I'm wondering how I preserve the
> functionality of these tools across the pending changes.
>    Any thoughts on how to provide a construction kit for command line/
> daemon users would be appreciated.

Dana Contreras
Twitter Platform Team

Reply via email to