But make sure you have a complete package for enduser (if you believe it's ready). This is very important esp. for PHP audience: download, unzip, upload, see "Hello World".
Automate this early, it will pay off fast.
Composer is a good Idea, but with first version we wanted to deliver one single package for the end users. We would use composer for the next releases.
defined('BASEPATH') OR exit('No direct script access allowed');
A PHP application can be dropped on any $2/month shared hosting account over FTP. That's the extent of the installation, and the host will support it for you. Even cheap unmanaged shared hosts will troubleshoot setting up PHP code for customers.
Ghost can only be set up by people that frequent sites like HN. A PHP app can be set up by anyone. That's why over 40 million people use WordPress where maybe 20k have used something like Jekyll.
For most people, I'm not sure that no FTP access is the dealbreaker it once was. I'm not saying that there's not validity in creating an alternative designed for PHP, but this criticism of Ghost doesn't ring particularly true to me for this reason: Ghost has had a total of four months on the market. Give it a year and see if the process is any easier. I'm going to say it might be.
There are lots of reasons to make a piece of software. Necessity is only a part of it, but even arguing from the position of an effort being unnecessary is short-sighted at best and belligerent at worst. People will make software even though similar offerings exist in the same space, and for the most part that's a good thing.
Your only other USP appears to be the ability to parse themes designed for a different engine. Is this going to be enough to distinguish Ospari from your competitors? Is it feasible for someone to write a WordPress plugin that achieves the same thing?