CapDB
Networked SQLite with replication.
Data
CapDB is networked SQL — pooled, secured, and replication-ready. It is a standalone SQL engine forked from SQLite3 that keeps the parts people love about SQLite (standard SQL, portable single-file databases, embeddability) and adds the production stack you usually have to bolt on later.
You can use it locally through the capdb shell or the embedded libcapdb library, or talk to it remotely over TLS with the capdb:// protocol. Built-in connection pooling with sane WAL defaults, a high-availability volume store, and primary/replica WAL-segment streaming mean the same engine scales from a desktop file to a replicated server.
Security is part of the design, not an afterthought: TLS transport, path jails, and generation fencing keep a networked database from becoming an open door. CapDB is also the optional networked backend for Capper, so the two projects share one data story.