An artifact is the output of an automated process that take some input and outputs artifacts. Its a generic term that can mean all kinds of things depending on the process and type of output. For example if you have a program that take an Open Office document in and produce a pdf and an epub file out then "pdf" and "ebup" would be the artifacts.

Artifact in this context is whatever is produced by build process. That is common convention in CI/CD context. And the base definition for "artifact" is very wide: anything artificial, as in not natural but made by humans.

Aren't the xhtml and TeX already artifacts though? They are produced from a script that parses the ODT

There is a bit of manual tweaking required in LaTeX that I haven't figured out how to automate yet. It's mainly related to chapter endings: if there are only 2 or 3 lines on the last page, I can subtly tighten the tracking from a paragraph on the second to last page and eliminate an almost-empty page. Also, I have PNG maps that currently need some one-off LaTeX directives to lay them out how I want. I could probably embed some properties in the ODT image, but that seems kind of janky. If I needed to regenerate from the ODT source, I use a Git three-way merge to update it (also janky, but it's expedient).

In software development, an artifact is a deployable file produced during the build process, such as a .jar, .zip, .exe, or Docker image.

In the publishing world, an artifact is something that is a product of processing code. e.g. the OP wants their code to generate files in various formats.