My personal search engine
For the last year and a half I’ve been running my own personal search engine. To be fair, it’s not actually doing any search indexing or searching. It’s really just a web endpoint that I can configure my browser to use as a search engine and is modeled after Meta’s internal bunnylol tool. In short, it has special handling for a series of simple commands or prefixes, and falls through to a Google Search if the query does not match one of these commands.
So, if I search for “event
And, again, if the query does not start with a known command/prefix it just redirects to a Google search for the query.
By configuring my browser to use this endpoint as my default search engine, my browser URL bar becomes a sort of command prompt.
The examples I have so far are quite simple and could honestly be implemented by a configuring Chrome’s search engines (which lets you pick a search engine based on a prefix) but in the future I’m looking forward to adding more sophisticated commands. For example:
short <some url>
could create a short link for the url passed and redirect to a page where I could copy itpaste <some text>
could create a paste containing “some text” and redirect to itevent Jan 2 at 6-7pm drinks with John
could parse the the date/time and create an event on my calendar for me (a Google Calendar feature I sorely miss)ga dau
could open Google Analytics with a view that show my side project’s daily active userstodo <some text>
could add an entry “some text” to my Notion todo list
This search engine is implemented as part of a larger personal server app I’ve written which includes:
Link shortener
Paste bin
Static image/file host
The server portion has a web UI as well as a CLI client which I install on all my computers. Maybe in the future I’ll write about the larger project. Let me know if you’re interested!