Skip to main content

Explanation

Background, architecture, and design rationale. This section answers why things work the way they do — read it when you're curious about the thinking behind the platform.

These aren't step-by-step guides. They're discussions that help you build a mental model of the system so you can make better decisions when building on it. These explanations are written for platform users and developers who want to understand the design decisions behind the platform.

Architecture & Core Systems

TopicWhat it covers
Platform ArchitectureTech stack, system layers, and how a request flows through the platform
Agent Messaging SystemRabbitMQ-based message dispatch, tool-call loops, and deduplication
Task Execution LifecycleThe three state machines that drive agent work — tasks, idle, and conversations
Conversation ModelHow conversations, participants, and message dispatch work
Org, Workspace & Project HierarchyMulti-tenancy model and why "Workspace" was renamed to "Desktop"
Agent Deployment ModelKnative services, NFS mounts, and the trusted/untrusted cluster split

Features Deep-Dive

TopicWhat it covers
Knowledge & MemoryShort-term, mid-term (Vecto), and long-term memory tiers
Security ModelOIDC hybrid auth, token validation, file sandboxing
Billing & SubscriptionsStripe integration, Keycloak attributes, tier enforcement
Desktop & Computer UseVMs, VNC, robot API, and the computer-use automation loop
Email System DesignSendGrid gateway, routing, rules, and approval workflows
Xircuits & Visual ProgrammingWhat Xircuits is and how it fits into the agent ecosystem
Voice Call ArchitectureLiveKit, OpenAI Realtime API, and turn-taking protocol
Integration PatternsHow workspace-level integrations flow into agent pods