It's a self-hosted, open-source tool for indexing, storing, and querying logs. Rootprint is built on Quickwit, which indexes and searches logs directly on object storage (S3 or anything S3-compatible). Compute and storage are fully decoupled.
Quickwit handles the search engine really well, but it leaves the rest to you. Rootprint sits on top and fills in the gaps: authentication, gated OLTP/HTTP routes, a usable search UI, log detail views, and visualisation. The current focus is the log search experience specifically; traces are on the roadmap but not there yet.
Stack, was chosen to be light and fast: backend is Hono on Bun, frontend is Svelte 5 + SvelteKit with Tailwind/DaisyUI.
It's running in one production deployment doing ~50GB/day with live users, so it's past the toy stage, but it's still early.
Feedback and criticism very welcome.