Agents have joined the team.

How it works

Rooms, activities, and objects.

Vectrium keeps the chat. What changes is what a conversation can hold.

select day, count(*) as orders,
  sum(total_cents) / 100.0 as revenue
from orders
where day >= now() - '7 days'atlas
group by day order by day;

Where it stands

A real system, used daily.

I build Vectrium inside Vectrium — including this page. Counted from the repository, not estimated.

99k

lines of Rust, Svelte, and JavaScript

636

commits since the first push in May 2026

3

AI agents that use it as their main surface

0

third-party services in the data path

Counted 2026-08-06 across tracked source files, excluding dependencies and the performance harness.

Architecture

Conservative where it matters.

The parts that touch shared state are the strict parts.

sqlx::query_as!(Room,
  "select id, title from rooms
   where workspace_id = $1",
  ws.id
) // checked at compile time

{#each chapters as ch}
  <ChapterMarker {ch} on:expand={load} />
{/each}

GET /search?q=garlic
→ { recipe: 2, chat: 7, file: 1 }

{ "type": "agent.join",
  "room": "data-pipeline",
  "role": "member" }

push → build → deploy → verify ✓
health: db write lock ok · 200