Hister - A private, full content search index that you control
23 points by Kratacoa
23 points by Kratacoa
I've seen the author @asciimoo reply in the comments here recently, yet the project has never had a dedicated submission.
Taking inspiration from the orange cousin, here it is .
I got quite excited about Hister, but was often annoyed by searches taking multiple seconds. Catering for "AI agents" with an MCP interface was the straw that made me uninstall.
Aw man, that turns me off of it too. Do you know of any potential alternatives?
I was hesitant to post this because I don't want to hijack this post in any way with self-promo, but I have a very potential alternative -- my project omry (that links to my personal forge; here is the codeberg mirror).
Full disclosure -- I have not used Hister, but from what I can tell, it is much more advanced in terms of UI (and other things), and it looks like it has a good web client. (I am currently working on a web client for omry, but it's not yet in a publicly usable state). Omry right now has only a terminal, REPL-style client that is somewhat barebones (but works).
One thing that Omry does that I think Hister doesn't is archive the full copy of the page, as well as extract the readable part for searching. Other than that, in general, Hister looks like it has a lot more features, and more polish (just being honest). But if all you want is to archive web pages and search them (with hopefully more to come later), and don't mind using the terminal to interact with it, you might find Omry useful! I am still actively working through many aspects, but having more users (and feedback) could really help me :) let me know if you try it!
also, for those who care, at the moment I have no plans to introduce LLM-related features in omry, and am generally cautious about LLMs. not meant to be a dig on Hister (I'm a bit jelly of how much further along it is tbh than where I wanted omry to be at this point); but I think asciimoto and I are trying to do different things.
Searches are faster than any external search engine for me. Mine runs on my local server. What about yours? That’s odd. I have a db with a little over 15k entires.
Also what about the MCP interface where can I find this?
I installed it a few days ago, and I've been running it since then to get a sense of it. I haven't needed to search it yet.
What's interesting to me is the extractors, but because there hasn't been a new release in a while those haven't helped me much. It's also not clear how it's going to work with social media, both on a post-by-post basis for SPAs, and more generally, for Mastodon sites, where the extractor model is more content-dependent than URL-based.
Still, it is a neat idea, and it's not using up a lot of CPU or space, so it's interesting to keep on.