{"product_id":"claude-multi-agent-routing-workflow-webhook-whatsapp-summary-n8n","title":"Claude Multi-Agent Routing Workflow: Webhook + WhatsApp Summary (n8n)","description":"\u003ch3\u003eTurn one request into a multi-step Claude “team” — with webhook delivery and a WhatsApp summary\u003c\/h3\u003e\n\u003cp\u003eThis \u003cstrong\u003en8n workflow\u003c\/strong\u003e accepts a task via \u003cstrong\u003eWebhook (POST)\u003c\/strong\u003e (or a manual run), uses \u003cstrong\u003eAnthropic Claude\u003c\/strong\u003e as a \u003cstrong\u003emulti-agent supervisor\u003c\/strong\u003e to route work across specialist agents (researcher, coder, analyst, writer, reviewer), iterates until completion, then returns the final result and sends a \u003cstrong\u003eWhatsApp summary\u003c\/strong\u003e.\u003c\/p\u003e\n\n\u003ch3\u003eWhat this workflow does\u003c\/h3\u003e\n\u003cul\u003e\n  \u003cli\u003e\n\u003cstrong\u003eReceives and validates input\u003c\/strong\u003e: A POST request hits an n8n \u003cstrong\u003ewebhook\u003c\/strong\u003e, and the workflow normalizes the payload into a task state with a \u003cstrong\u003estep counter\u003c\/strong\u003e and \u003cstrong\u003ework history\u003c\/strong\u003e. If the task is empty, it immediately returns a \u003cstrong\u003e400 webhook response\u003c\/strong\u003e.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eSupervises routing with Claude\u003c\/strong\u003e: The workflow builds a supervisor prompt and calls the \u003cstrong\u003eAnthropic Messages API (Claude)\u003c\/strong\u003e to decide the \u003cem\u003enext specialist route\u003c\/em\u003e and instruction.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eCalls exactly one specialist per step\u003c\/strong\u003e: It routes to a single Claude call—\u003cstrong\u003eresearcher\u003c\/strong\u003e, \u003cstrong\u003ecoder\u003c\/strong\u003e, \u003cstrong\u003eanalyst\u003c\/strong\u003e, \u003cstrong\u003ewriter\u003c\/strong\u003e, or \u003cstrong\u003ereviewer\u003c\/strong\u003e—to generate output based on the supervisor’s instruction.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eIterates until FINISH (with guardrails)\u003c\/strong\u003e: Results are appended to a shared work log, the step counter increments, and routing repeats until Claude signals \u003cstrong\u003eFINISH\u003c\/strong\u003e or a guardrail ends the run (including loop detection and error fallback).\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eSynthesizes and responds\u003c\/strong\u003e: After completion, the accumulated specialist work is sent to Claude for \u003cstrong\u003efinal synthesis\u003c\/strong\u003e. The workflow returns the \u003cstrong\u003efinal answer\u003c\/strong\u003e and \u003cstrong\u003erouting trace\u003c\/strong\u003e in the webhook response, and sends a \u003cstrong\u003etruncated summary\u003c\/strong\u003e via \u003cstrong\u003eWhatsApp\u003c\/strong\u003e.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3\u003eUse cases\u003c\/h3\u003e\n\u003cul\u003e\n  \u003cli\u003eSaaS operators: turn customer requests into researched, structured outputs with automated review.\u003c\/li\u003e\n  \u003cli\u003eAutomation engineers: build Claude-powered “agent loops” with routing visibility (trace) and controlled step limits.\u003c\/li\u003e\n  \u003cli\u003eInternal teams: generate drafts (writer), verify them (reviewer), and produce final deliverables in one workflow run.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3\u003eTechnical details\u003c\/h3\u003e\n\u003cul\u003e\n  \u003cli\u003e\n\u003cstrong\u003eTrigger\/HTTP\u003c\/strong\u003e: n8n \u003cstrong\u003eWebhook\u003c\/strong\u003e (POST) + optional manual trigger.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eRouting \u0026amp; control\u003c\/strong\u003e: \u003cstrong\u003eif\u003c\/strong\u003e, \u003cstrong\u003eset\u003c\/strong\u003e, \u003cstrong\u003ecode\u003c\/strong\u003e, \u003cstrong\u003eswitch\u003c\/strong\u003e for task state, guardrails, and looping logic.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eLLM integration\u003c\/strong\u003e: \u003cstrong\u003eAnthropic HTTP Request\u003c\/strong\u003e using an \u003cstrong\u003ex-api-key\u003c\/strong\u003e (Claude Messages API).\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eMessaging\u003c\/strong\u003e: \u003cstrong\u003eWhatsApp\u003c\/strong\u003e node to send a truncated summary.\u003c\/li\u003e\n\u003c\/ul\u003e","brand":"N8N Commerce","offers":[{"title":"Default Title","offer_id":46123136876723,"sku":"N8N-20120","price":36.99,"currency_code":"GBP","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0749\/6279\/6723\/files\/w7UExzWNOERi89AMHSBYi_Mqj6fnVK.png?v=1790672908","url":"https:\/\/buyflowscripts.com\/products\/claude-multi-agent-routing-workflow-webhook-whatsapp-summary-n8n","provider":"N8N Commerce","version":"1.0","type":"link"}