Concepts

Deep dives into ark-operator’s architecture and runtime primitives — how agents work, LLM providers, task queue, MCP servers, observability, and more.

Deep dives into how ark-operator works under the hood.

PageDescription
How It WorksThe reconcile loop, agent pods, task flow, and mental model
ProvidersAnthropic, OpenAI, Ollama, and custom LLM providers
Task QueueRedis Streams internals and pluggable backends
MCP ServersModel Context Protocol tool server integration
MemoryIn-context, Redis, and vector-store memory backends
ObservabilityOTel traces, metrics, and audit events
ScalingManual scaling, kubectl scale, and daily token budget scale-to-zero
Semantic Health ChecksLLM output validation via /readyz