QuestionQ43

Agentic Architecture & Orchestration

The coordinator agent has AgentDefinitions configured for each of the four specialized subagents, with suitable descriptions, prompts, and tool restrictions. During testing, the coordinator correctly identifies when delegation is needed—it produces messages such as “I’ll ask the web search agent to find sources on this topic”—but no subagent execution ever takes place. The coordinator then acts as though delegation occurred and continues using incomplete information. Logs show no errors. What is the most likely cause?

Explanation

AgentDefinitions make subagent types available for configuration, but delegation requires the coordinator to invoke the Task tool. When Task is not included in the coordinator’s allowed-tools configuration, it cannot dispatch work to a subagent even if its reasoning identifies that delegation would be appropriate. Tool permissions control whether a configured tool may be invoked.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!