engyossefyossry-crypto/autonomous-devsecops-triage-engine-pro

This project is an automated incident response system built on top of n8n, a visual workflow automation tool. When something goes wrong in a production system, this workflow receives the alert, uses several AI models to analyze it, and routes the results to the right people and tools, such as Slack, email, Discord, GitHub, and Telegram. The README describes this as an upgrade from an earlier, simpler version of the same idea. Instead of trusting one AI model to handle everything in a straight line, this version splits the work across several different AI models, each assigned to a specific job. One model acts as a supervisor that checks other models' work for consistency and makes sure responses follow a strict format. Another model focuses on reading error messages and stack traces to suggest code fixes. A separate model is responsible for judging security risk, and yet another is responsible for turning technical details into plain language status updates for humans. The workflow is split into two paths. A success path runs multiple AI checks in parallel, then merges and audits their answers before updating a Grafana dashboard, a tool for visualizing system health data. A separate failure path exists specifically to catch problems like API rate limits or network timeouts, so that an infrastructure failure gets logged and reported instead of silently crashing the whole pipeline. The system also keeps a written record of what happened during each incident, including which file and line number was involved and what code fix, if any, was suggested, stored directly inside n8n's own data tables rather than an external database. This is aimed at teams already using n8n who want to add AI assisted triage to their incident response process, and it assumes familiarity with workflow automation tools, API keys for multiple AI providers, and basic DevOps concepts like rate limits and webhooks.