Implemented agent streaming

README.md

@@ -123,7 +123,7 @@ Example profile: src/test/resources/llama3.3-70b-azure.properties
 
 Client service: clientService.getClient(ModelCategory.MEDIUM);
 Client usage details: CompletionClientTest and AzureAiClientTest
-Agentic flow: AgentTest#agenticFlow  
+Agentic flow: AgentTest#agenticFlow and AgentTest#streamedAgenticFlow
 
 ## Completions and Streaming
 
@@ -143,10 +143,6 @@ You can intercept every Conversation and Completion by implementing CompletionIn
 This is for example useful for accounting purposes or to implement a circuit breaker.
 The client will autowire all implementations and intercept requests accordingly.
 
-# Agentic Flows
-
-This library contains a graph executor for implementing comprehensive processes using agent-based design patterns. See class AgentTest for examples.
-
 # Grader
 
 The library includes a grader component that evaluates the correctness of agent responses.
@@ -155,6 +151,30 @@ Example:
 
     assertTrue(Grader.assessScoreBinary("Does the following blog post mention the feature " + feature, resultContext.getStringValue("blogPost"), this.clientService));
 
+# Agentic Flows
+
+This library contains a graph executor for implementing comprehensive processes using agent-based design patterns. 
+
+You can implement your Agents either by creating an Agent record or by extending BaseAgent - choose the approach that best fits your needs.
+The agentic flow is started with an AgentContext and an initial agent.
+Example:
+
+    final var context = new AgentContext(List.of(new Agent1(), new Agent2()));
+    context.getContext().setStringValue("key", "Payload here");
+    context.getContext().setStringValues("key2", List.of("All context keys", "support collections", "for aggregation purposes"));
+    context.getContext().setValue("key3", new RandomDatatype());
+    final var resultContexts = agentExecutorService.walk("Agent1", context)
+    final var resultContext = resultContexts.getFirst(); // You may receive multiple contexts if your graph branches out
+    LOGGER.info(resultContext.getStringValue("blogPost") + "\n\n" + resultContext.getGraph().toString() + "\n" + resultContext.getRequestStatistic().toString());
+
+Streaming is also supported.
+
+See class AgentTest for examples.
+
+# Agent implementations
+
+
+
 # Announcement generated by AgentTest#agenticFlow
 
 ## Introducing Advanced Agent Orchestration: Dynamic Conversations & Parallel Processing