About the Project: Simplifying Network Operations

The proposed solution aims to significantly improve the network monitoring administrator experience through Artificial Intelligence (AI). It represents an AIOps (AI for IT Operations) approach, transforming the complex interaction with the existing supervision system (Zabbix) into a simple, intuitive, and effective dialogue via WhatsApp.

AI and Automation Features

• Conversational WhatsApp Interface: Allows administrators to access monitoring information through an intuitive WhatsApp interface, providing real-time mobile access to network metrics.
• Secure Authentication: Access to secure Zabbix commands is protected by chatbot authentication. The system temporarily blocks access after 3 incorrect password attempts.
• Natural Language Analysis (`!zabbix analyze`): Provides a natural language summary of system performance, identifies priority issues, and offers recommendations using the Gemini API to interpret raw JSON data from Zabbix.
• Failure Prevention (`!zabbix predict`): Estimates the risk of an incident for a host based on its metrics (RAM, CPU, ROM) using the Random Forest machine learning algorithm.
• Automated Ticket Generation: The workflow automates the creation of support tickets via the GLPI API as soon as a technical incident is detected by Zabbix.
• Enriched Notifications: Alerts received via WhatsApp are detailed and structured, including the GLPI ticket ID, a direct link, the host name, problem description, severity level, and a non-technical analysis with solution recommendations provided by the AI.