This was my learning repository for many years and it contains many early examples of automation and other projects. Please note that there is a new repository that has all the current scripts to automate documentation, the Confluence documentation release, and so on. There is also a separate repository for the API examples project.
This repository contains many scripts that process code or API responses or other files to create reference documentation. The first automation I created was a shell script that contained AWK commands that read code. For example, a Java API errors file with AWK to process API Error file. Now for this process I use a Python script that reads the API errors generated in the build and creates a table in the Abiquo wiki.
Another early example is the documentation from the central configuration properties file. The Development team commits to the central file, and the scripts then process it for the wiki. See AWK to process properties file. The latest version is in Python and it manages multiple variants of single properties.
A common customer use case is to create a tenant in the cloud platform. The customer will then automate this action with the API for their specific environment. I created a very quick Python script to test the steps in the customer use case. I later also created a how-to to describe the steps: how-to create a tenant via API.
When I write an API how-to, I think it's sometimes helpful to check the process with code, not just with cURL commands. I enjoyed using Python to implement how to add a NAT rule via API. The Python script is very basic, because it is just to test the process. See Add a NAT rule to a VM with Python.
I had an HTML version of the API docs and I used Python to scan it for missing data. In some cases, the privileges required to use a request were not appearing in the API docs. Rather than check 100 pages containing 700 methods by hand, I wrote a script. See Check roles or privileges in API docs
Migrating from Confluence Server to Confluence Cloud took several months. When you get to Cloud, your pages will be in the "legacy editor". Hopefully, you will be able to easily convert them to the "new editor" and it will offer all of the functionality necessary for your documentation use case. As part of a migration, I wrote scripts to automate many tasks, such as, check which editor the page is using.
There were some new UI screens that displayed custom metrics. However, custom metrics are metrics that users have added with the API or a tool such as collectd running on the VM. To quickly get some screenshots with data, I wrote a very rough script to push custom metrics.