orchestrator

Purpose

This skill handles multi-instance delegation in distributed systems, routing tasks to instances A-H, managing fanout and aggregation of responses, spawning/sending sessions, and performing health checks. It's designed for coordinating complex workflows across distributed-comms clusters.

When to Use

Use this skill for scenarios involving multiple instances, such as load balancing requests across A-H, aggregating data from distributed nodes, managing long-running sessions, or ensuring instance health in fault-tolerant systems. Apply it when single-instance processing isn't sufficient, like in scalable web services or parallel computations.

Key Capabilities

• Routing: Direct tasks to specific instances (e.g., A-H) using targeted delegation.
• Fanout/Aggregation: Broadcast requests to multiple instances and collect/aggregate responses, supporting operations like parallel queries.
• Session Management: Spawn new sessions with sessions_spawn and send data via sessions_send, enabling stateful interactions.
• Health Checks: Periodically verify instance status with endpoints like /api/health/check, returning JSON with status codes (e.g., 200 for healthy).
• Error Resilience: Automatically retry failed delegations up to 3 times based on configurable thresholds.