ReAcTree Agent Coordination Patterns

What problem does it solve?

This Skill provides a structured approach to decomposing complex tasks into hierarchical agent trees and control-flow nodes, enabling robust reasoning and parallelism.

Core Features & Use Cases

• Hierarchical Agent Trees: complex tasks decomposed into tree structures where each node can be an agent (reasoning + action) or control flow coordinator.
• Control Flow Nodes: Six coordination patterns: Sequence, Parallel, Fallback, LOOP, CONDITIONAL, TRANSACTION.
• Dual Memory Systems: Working Memory and Episodic Memory to cache verifications and learn from past subgoals.
• Rails-ready patterns: adaptable to Rails service objects and orchestration layers.

Quick Start

Define a small task broken into a root Sequence node with two child tasks, one for data retrieval and one for processing, and observe how the coordinating node triggers subgoals and memoizes results.