I wrote a BASH script for automated downloading of selected arXiv RSS feeds.

That content is parsed into two documents:

* keyword-matched articles of interest;
* the remaining articles.

The script can be scheduled to run daily via crontab, or manually executed.

By example, today (Jan 10, 2019) among five arXiv RSS feeds my RSS reader provided 690 entries (including duplicated, cross-posted entries) while my script returned 344 de-duplicated (unique) articles.

* script: https://persagen.com/files/misc/arxiv-rss.sh
* accompanying research blog post: https://persagen.com/2019/06/10/arxiv-rss.html