Glossary

Tenant isolation

Strict separation so one customer's data and agents can never reach another's.

category ▸ Agents & Identity

audit/verify
#101hash ✓#102hash ✓#103hash ✓#104hash ✓#105hash ✓
head 0x9f3a…c1ok: true
verifyChain ▸ chained SHA-256 · signed receipts
What it means

Tenant isolation, in plain language.

Tenant isolation is the guarantee that, in a multi-tenant system, each customer's data, identities, and activity are walled off from every other customer's. No query, agent, or action belonging to one tenant can read or affect another, even by mistake.

For regulated buyers, isolation is non-negotiable: it is the foundation of confidentiality and the precondition for any compliance claim. Isolation that is enforced structurally — built into every data access — is far stronger than isolation that depends on application code remembering to filter.

In Cortex

How Cortex implements it.

This term isn't abstract here — it maps to a real capability in the runtime. Here is exactly how Cortex enforces or relates to it.

Agents & Identity

Cortex keys data on (tenantId, id) throughout, so tenant scope is part of every lookup rather than an afterthought. Agent identities, ontology objects, policies, ledger records, and budgets are all tenant-scoped, and cross-tenant transfer only happens out-of-band via a signed pack artifact — never a direct cross-tenant write.

For the strictest deployments, an air-gapped or single-tenant VPC install runs the full runtime with local models, so data never leaves the customer's perimeter at all.

See Tenant isolation enforced, not just defined.

Book a walkthrough and watch the controls in this glossary return real verdicts, seal real evidence, and trace every fact back to its source.