A programming framework for agentic AI
Update architecture.md (#588)
gagb7f25d28aacbc
diff --git a/python/packages/autogen-core/docs/src/user-guide/core-concepts/architecture.md b/python/packages/autogen-core/docs/src/user-guide/core-concepts/architecture.mdindex df2c9af95..e3580f385 100644--- a/python/packages/autogen-core/docs/src/user-guide/core-concepts/architecture.md+++ b/python/packages/autogen-core/docs/src/user-guide/core-concepts/architecture.md@@ -1,32 +1,31 @@-# AGNext Architectures+# Agent Runtime Environments-AGNext is a framework for building multi-agent applications with AI agents.-At the foundation level, it provides a runtime envionment to facilitate-communication between agents, manage their identities and lifecycles,+At the foundation level, the framework provides a _runtime envionment_, which facilitates+communication between agents, manages their identities and lifecycles,and enforce security and privacy boundaries.-AGNext supports two types of runtime architectures: standalone and distributed.-Both architectures provide a common set of APIs for building multi-agent applications,-so you can switch between them with no change to your agent implementation.-Each architecture can also have different implementations.+It supports two types of runtime environment: *standalone* and *distributed*.+Both types provide a common set of APIs for building multi-agent applications,+so you can switch between them without changing your agent implementation.+Each type can also have multiple implementations.## Standalone Agent RuntimeStandalone runtime is suitable for single-process applications where all agentsare implemented in the same programming language and running in the same process.-In Python API, an example of standalone runtime is the {py:class}`~autogen_core.application.SingleThreadedAgentRuntime`.+In the Python API, an example of standalone runtime is the {py:class}`~autogen_core.application.SingleThreadedAgentRuntime`.-The following diagram shows the standalone runtime architecture of AGNext.+The following diagram shows the standalone runtime in the framework.-+-Agents communicate via messages through the runtime. The runtime manages-the lifecycle of agents.+Here, agents communicate via messages through the runtime, and the runtime manages+the _lifecycle_ of agents.Developers can build agents quickly by using the provided components including-routed agent, AI model clients, tools for AI models, code execution sandboxes,+_routed agent_, AI model _clients_, tools for AI models, code execution sandboxes,model context stores, and more.-They can also implement their own from scratch, or use other libraries.+They can also implement their own agents from scratch, or use other libraries.## Distributed Agent Runtime@@ -34,17 +33,17 @@ Distributed runtime is suitable for multi-process applications where agentsmay be implemented in different programming languages and running on differentmachines.-+A distributed runtime, as shown in the diagram above,-consists of a host servicer and multiple workers.+consists of a _host servicer_ and multiple _workers_.The host servicer facilitates communication between agents across workersand maintains the states of connections.-The workers run agents and communicate with the host servicer via gateways.+The workers run agents and communicate with the host servicer via _gateways_.They advertise to the host servicer the agents they run and manage the agents' lifecycles.Agents work the same way as in the standalone runtime so that developers can-switch between the two runtime architectures with no change to their agent implementation.+switch between the two runtime types with no change to their agent implementation.## API Layers