Diagnostic Slackbot

The Diagnostic Slackbot is an AI-powered Slack bot that automates infrastructure diagnostics. When an incident occurs or a user reports an issue, the bot can analyze WAF logs, Kubernetes events, and system metrics to provide actionable diagnostic information directly in Slack, reducing the time from alert to understanding.

The bot uses curated, tunable prompts to ensure that AI-generated analysis is relevant, accurate, and aligned with the organization’s infrastructure. Rather than generic AI responses, the prompts are tailored to the specific systems and failure modes that matter to the team. Operators can adjust the prompts to improve diagnostic quality over time.

Users interact with the bot through natural Slack conversations. They can ask about specific incidents, request analysis of particular systems, or trigger broader diagnostic sweeps. The bot correlates data across multiple sources to build a coherent picture of what happened and why.

This approach puts diagnostic capability in the hands of the broader team without requiring everyone to know how to query Loki, navigate Grafana dashboards, or parse Kubernetes events directly. It democratizes incident investigation while maintaining the depth that experienced operators expect.

GitHub: https://github.com/nikogura/diagnostic-slackbot