What problem does it solve? Building systems where multiple AI agents coordinate is error-prone: routing correctness, state integrity, and failure isolation are hard to retrofit. This Skill provides architectural patterns for agent registries, job queues, LLM fallback routing, state machines, tool dispatch authorization, and prompt injection defenses so these problems are solved by design. ## Core Features & Use Cases - Agent Registry & Contracts: Define each agent role with an explicit model, fallback model, tool allowlist, timeout, and versioned system prompt before writing any code. - BullMQ Job Architecture: Structure stateless agent jobs with correlation IDs, retry policies, and separate Redis connections for queues, workers, and events. - LLM Router with Fallback: Wrap every model call with input sanitization, timeouts, and automatic fallback to a secondary model on rate limits or timeouts. - Use Case: When building a pipeline where an orchestrator decomposes a task, dispatches sub-tasks to coder and researcher agents, evaluates results, and synthesizes a final answer, use this Skill to design the state machine, queue topology, and observability spans before implementation. ## Quick Start Ask the AI to design the agent registry, BullMQ job chain, and LLM fallback routing for your multi-agent system before writing any agent code.
