Neruva holds real working notes, so the basics are covered. This page says exactly what is in place today, and nothing more.
The API runs on Google Cloud Run in the us-central1 region (Iowa, USA). Stored data sits on Google Cloud, which encrypts data at rest by default.
Every request goes over HTTPS and carries an API key. A key is tied to one tenant, and requests can only read and write that tenant's namespaces. Keys are issued by me, by hand.
It is stored, indexed and returned when asked for. It is not used to train any model, mine or anyone else's, and it is not sold or shared.
Records can be forgotten by kind, tag or time, and an entity can be removed from the knowledge graph completely. A whole namespace can also be exported first.
Neruva is a one-person prototype. There is no SOC 2 report, no formal audit and no uptime promise. If that matters for your use, tell me and we can talk about what would be needed.
Email kyle@simgen.dev with anything that looks wrong. I will reply directly.