A [[https://www.lesbonscomptes.com/recoll/][cross-platform desktop search engine]] that can search through [[id:ccb3bc14-a801-4ed0-b066-50b1bcd853aa][File metadata]] from various formats including HTML, text files, PDF, and more.
It can be automated easily with the command line and it has a GUI available as well as various third-party frontends.
Moreover, it has a Python 3 API that can extend Recoll for support to additional file formats, for example.
Like most web search engines, it searches through an index which has to be generated first (i.e., ~recollindex~).
With the file changes, the index has to be regenerated.