Swarm Orchestration

What problem does it solve? Coordinating multiple AI agents on complex tasks requires manual management of task distribution, shared state, failure handling, and execution ordering, which becomes unmanageable beyond a single agent. ## Core Features & Use Cases - Topology Management: Initialize mesh, hierarchical, or adaptive swarm topologies with configurable agent counts and communication patterns. - Task Orchestration: Execute tasks in parallel, sequential pipelines with dependencies, or fully automatic mode with time and agent constraints. - Shared Memory & Resilience: Coordinate agents through shared swarm memory, load balancing, retry policies, and performance metrics. - Use Case: A developer building a REST API spawns coder, tester, and reviewer agents, runs implementation and test-writing in parallel, and shares the API schema through swarm memory so all agents stay synchronized. ## Quick Start Initialize a mesh swarm with five agents using agentic-flow, then spawn coder, tester, and reviewer agents and orchestrate the task of building a REST API with tests in parallel mode.