System
System endpoints are grouped into four areas. Organizations is where you manage organizations and the users inside them. Models manages the AI providers and model versions Cortex can call. Billing covers the plan catalog and per-organization subscription, billing profile, payment method, and invoice lifecycle. Runs gives read-only visibility into cross-org run activity for observability tooling.
Organizations
Organization records and the users that belong to them. CRUD on orgs, lifecycle (suspend, soft-delete, restore), and per-org user management.
Organizations, organization members.
Models
AI providers configured in the system — including each provider's encrypted API key, live model listings from each vendor's own API, version import and refresh — and the model categories and versions Cortex knows how to call.
AI providers, model categories, model versions.
Billing
Plan catalog, per-organization subscription lifecycle, billing profiles, saved payment methods, and invoices across every organization. Stripe-source rows arrive via webhook (when wired); manual operator-managed rows live here too via the same endpoints. Cards are held by Stripe — these endpoints store only the payment method reference.
Plans, subscriptions, billing profiles, payment methods, invoices.
Runs
Read-only, cross-org run activity for the console's observability surface: list runs with status, stuck, and search filters, and inspect a single run's full act DAG.
Runs, run acts.
Platform
One aggregate for the console Dashboard: runs per day and week-over-week, tenant and user counts, subscriptions with an MRR estimate, open and overdue invoices, wallets, platform-wide AI spend, and the busiest organizations.
Platform overview, top organizations, spend by provider, runs and metric series.