tenancyjs-identifiers

Framework-neutral tenant extraction and resolution outcomes for TenancyJS.

New to TenancyJS? Start with the docs → — install, the tenancyjs-cli CLI, and how this package fits with an adapter + integration.

Resolvers extract identifiers from untrusted request metadata. A TenantStore does the registry lookup, and TenantResolutionChain returns an exhaustive, fail-closed outcome. Resolving proves a tenant exists — it never proves the user may act as it, so the chain requires you to decide membership: pass an authorize hook, or opt out with trustResolution for a source you trust.

const chain = new TenantResolutionChain({
  resolvers: [
    new HeaderTenantResolver(),
    new SubdomainTenantResolver({ centralDomain: "app.example.com" }),
  ],
  store,
  // Verify the authenticated principal may act as the resolved tenant.
  authorize: ({ tenant, principal }) => principal.teamIds.includes(tenant.id),
});

const outcome = await chain.resolve(
  { host: "acme.app.example.com", headers: { "x-tenant-id": "tenant_acme" } },
  { principal: currentUser },
);

A raw header is spoofable, so trustResolution cannot be combined with a header/host resolver — it must go through authorize. Resolver order is precedence; a present higher-priority identifier that is invalid or unknown fails closed and never falls through.