Unified Orchestration Layer
Three authoring APIs. One compiled intermediate representation. A single runtime that executes every graph with streaming, checkpointing, guardrails, memory, and capability discovery built in.
mission() API
Live run examples/mission-api.mjs.
Agency Collaboration
Live runs agency-graph.mjs, agency-streaming.mjs, emergent-hierarchical-spawning.mjs.
workflow() DSL
The right way to ship a multi-step agent is rarely the most flexible way. A graph that can loop and replan is great when you don't know what shape the work takes; it's overkill — and harder to operate — when you do know. workflow() exists for the second case. You declare the steps in the order they run, the compiler builds the execution graph, and a static cycle check rejects anything that would loop. The output is the same CompiledExecutionGraph the cyclic builders produce, so you can swap a workflow() for an AgentGraph later without re-architecting the runtime around it.
Agent Communication
AgentCommunicationBus — structured messaging between AgentOS GMIs in a multi-agent agency. Point-to-point, broadcast, request/response, pub/sub, handoff, threading. Three layers: Message Router → Subscription Manager → Delivery Manager with retry, ACK, and audit.
AgentGraph
When workflow() is too rigid and mission() is too far ahead of where the runtime currently plans, the answer is AgentGraph — explicit node and edge construction with cycles, conditional routing, subgraph composition, and the discovery and personality edges that don't exist in any other open agent framework. It compiles to the same CompiledExecutionGraph IR as the higher-level builders, but it gets you full control over the topology before compilation.
Checkpointing and Time-Travel
The AgentOS Unified Orchestration Layer has built-in support for checkpoints, resume after failure, and time-travel debugging via the ICheckpointStore interface.
Planning Engine
PlanningEngine — multi-step task planning for AgentOS agents. ReAct, Tree-of-Thoughts, Least-to-Most decomposition, Reflexion self-correction, autonomous loops, checkpoint/rollback, agency distribution across GMIs.
Human-in-the-Loop
Overview