It's somewhat structured; I use both broad categories and a tag system. I can also add additional comment text to archived pages. It's all patched together with shell scripts and some Lua (since that's what I'm familiar with). `ripgrep` is the utility used for searching. It's fast enough for me even when I don't use any kind of category filtering, but I have a beefy computer and use NVMe drives, so YMMV.