n8n Webhook Self-Healing for Failed Workflows w/ AI API

Automatically self-heal failed n8n workflows with AI-driven recovery strategies

This n8n workflow listens for failed-execution webhook events, analyzes the error, consults an external AI HTTP API to generate recovery plans, attempts an automated fix, and returns a structured self-healing report to the caller.

What this workflow does

• Receives failure details via webhook: Accepts a POST request with context such as the failing workflow/node and errorMessage/errorCode (webhook path: /self-heal).
• Validates and classifies the failure: Checks the payload, extracts workflow + node + error + retry count, then classifies severity and retryability.
• Fetches execution history: Calls an HTTP endpoint to retrieve recent execution history and related error patterns for the failing workflow/node.
• Generates ranked recovery strategies with AI: Sends the failure context + history to an external AI HTTP endpoint and receives ranked recovery strategies.
• Executes a primary recovery action: Triggers an automated recovery step (e.g., retry, fallback, or escalation) via another HTTP endpoint.
• Logs the outcome and responds: Sends the result to an outcome logging HTTP endpoint, then returns a consolidated self-healing report in the webhook response.

Use cases

• Speeding up recovery when an automation engineer notices repeated node failures and wants automated retry/fallback behavior.
• Reducing incident time for SaaS operators by turning failed workflow executions into an AI-assisted recovery workflow.
• Capturing consistent post-recovery reporting for troubleshooting and monitoring across multiple n8n automations.

Technical details

• n8n nodes: code, wait, webhook, sticky note, http request, respond to webhook.
• External services (HTTP endpoints): execution history fetch, AI strategy generation, recovery execution, and outcome logging.
• Setup note: Replace placeholder httpbin.org URLs with your real internal endpoints (add HTTP authentication if required).