supermemory/apps/docs/concepts/multi-tenancy.mdx
Dhravya Shah f882f1104d
docs: restructure documentation site and update integration UI (#1331)
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
2026-07-21 20:19:30 -07:00

114 lines
4.8 KiB
Text

---
title: "Multi-tenancy Overview"
sidebarTitle: "Overview"
description: "How Supermemory isolates and organizes memories across users, tenants, and projects"
icon: "users"
---
Most apps built on Supermemory serve more than one user, customer, or tenant out of a single Supermemory organization. Multi-tenancy is how you keep those memories apart — so User A's data is never visible to User B, and so you can still slice and query within a user's own data by things like category, status, or date.
Supermemory gives you two complementary tools for this:
<CardGroup cols={2}>
<Card title="Container Tags" icon="folder" href="/concepts/container-tags">
**Isolation.** A container tag is a hard boundary — its own namespace. Memories in one tag are never returned by a search scoped to another tag.
</Card>
<Card title="Metadata Filtering" icon="database" href="/concepts/filtering">
**Organization.** Metadata is a set of custom key/value properties on a memory that you filter by — category, priority, date, participants, anything you define.
</Card>
</CardGroup>
They solve different problems, and most production apps use both together.
---
## Why two mechanisms
It's tempting to reach for one tool and make it do everything, but tags and metadata aren't interchangeable — they answer different questions.
| Question | Answer |
|----------|--------|
| "Which tenant does this memory belong to?" | **Container tag** |
| "Within this tenant's memories, which ones match `status: open`?" | **Metadata filter** |
| "Can this API key even see tenant X's data?" | **Container tag** (enforced as an access boundary) |
| "Find memories tagged `engineering` created after March" | **Metadata filter** |
A container tag decides **whether a memory is reachable at all** for a given request. Metadata decides **which of the reachable memories match**. Filtering never crosses a container tag boundary — you can't use metadata to peek into another tenant's container.
---
## How they work together
A typical multi-tenant write scopes the memory to a tenant with a container tag, then attaches metadata for finer-grained querying later:
```typescript
await client.add({
content: "Customer requested a refund for order #4821",
containerTag: "org_acme", // isolates to the "acme" tenant
metadata: {
category: "support",
status: "open",
priority: "high",
},
});
```
And a search combines both: the container tag restricts *which tenant's data* is in scope, and filters narrow down *which memories within that tenant* come back:
```typescript
const results = await client.search({
q: "refund request",
containerTag: "org_acme",
searchMode: "documents",
filters: {
AND: [
{ key: "category", value: "support" },
{ key: "status", value: "open" },
],
},
});
```
<Note>
Container tags are **required** for isolation and validated as an access boundary. Metadata filters are **optional** — a search with just `containerTag` and no `filters` still only returns that tenant's memories.
</Note>
---
## Choosing your boundary
Container tags are the layer that should map to your actual tenancy model — pick the level that matches what "one isolated space" means in your app:
| Pattern | Example | Use case |
|---------|---------|----------|
| Per-user | `user_{userId}` | Consumer app, personal memory per user |
| Per-tenant/org | `org_{orgId}` | B2B SaaS, one container per customer org |
| Hierarchical | `org:{orgId}:user:{userId}` | Multi-level — isolate by org, and optionally drill into a user within it |
| Per-project | `project_{projectId}` | Workspace- or project-scoped content |
Everything *within* that boundary — categories, statuses, dates, custom fields — is metadata, not a new tag. Don't create a new container tag for every property you want to filter on; that's what metadata is for.
---
## Access control
Container tags aren't just organizational — they're enforced as an authorization boundary. API keys and org members can be restricted to specific tags, so a request for a tag outside the caller's allowed set is rejected with `403 Forbidden` rather than silently filtered. See [Container Tags → Access control](/concepts/container-tags#access-control) for the details.
---
## Next steps
<CardGroup cols={2}>
<Card title="Examples" icon="list-checks" href="/concepts/multi-tenancy-examples">
Personal agents, company agents, email assistants, and support platforms.
</Card>
<Card title="Container Tags" icon="folder" href="/concepts/container-tags">
How isolation works, naming rules, and access control.
</Card>
<Card title="Organizing & Filtering" icon="filter" href="/concepts/filtering">
Metadata filter types, combining `AND`/`OR`, and query limits.
</Card>
<Card title="Scoped API keys" icon="key" href="/authentication#scoped-api-keys">
Mint keys that can only touch one container tag.
</Card>
</CardGroup>