As mentioned in a previous submission I've built a daemon to automate monitoring topics on Hackernews. It takes a list of queries (say, "Ruby", "C#" or "Laravel" for example) and exposes stories that match them through an RSS feed.
In the two major versions released since then, I’ve made several improvements:
* added support for Lobste.rs
* switched from scraping Hackernews' HTML pages to using the official API, as suggested by another HN user
* replaced the JSON file-based storage with SQLite
Let me know what you think, and whether this would be useful for you.