Add agentic module: sequential, parallel, loop, supervisor and error-recovery workflows

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01G8ikz8xdWuTP5yun8DZ1hk

agentic/output/07-name-collision.txt

@@ -0,0 +1,32 @@
+Two agents, both with a method named find. What the supervisor model was shown:
+
+SYSTEM: SystemMessage { text = "You are a planner expert that is provided with a set of agents.
+You know nothing about any domain, don't take any assumptions about the user request,
+the only thing that you can do is rely on the provided agents.
+
+Your role is to analyze the user request and decide which one of the provided agents to call next to address it.
+You return an agent invocation consisting of the name of the agent and the arguments to pass to it.
+
+If no further agent requests are required, return an agentName of "done" and an argument named
+"response", where the value of the response argument is a recap of all the performed actions,
+written in the same language as the user request.
+
+Agents are provided with their name and description together with a list of applicable arguments
+in the format {'name', 'description', [argument1: type1, argument2: type2]}.
+
+Decide which agent to invoke next, doing things in small steps and
+never taking any shortcuts or relying on your own knowledge.
+Even if the user's request is already clear or explicit, don't make any assumptions and use the agents.
+Be sure to query ALL necessary agents.
+
+The comma separated list of available agents is: '{'find$0', 'Finds a train', [destination: String]}, {'find$1', 'Finds a bus', [destination: String]}'.
+
+
+", attributes = {} }
+USER: UserMessage { name = null, contents = [TextContent { text = "The user request is: 'Get me to Lisbon'.
+The last received response is: ''.
+
+You must answer strictly in the following JSON format: {
+"agentName": (type: string),
+"arguments": (type: java.util.Map<java.lang.String, java.lang.Object>)
+}" }], attributes = {} }