VersusControl/versus-incident ? reverse-engineered prompt
Reverse engineered prompt
Build me a self hosted incident monitoring app that acts like an AI SRE teammate.
It should watch my logs, learn what normal looks like, and only open an incident when something new or unexpected shows up. I also want a simple webhook endpoint so tools like Alertmanager, Grafana, Sentry, or my own scripts can send alerts into the same system.
Make it run on port 3000, include a basic admin dashboard where I can review what the agent has learned, and let me turn the agent on in training, shadow, or detect mode. It should send incidents out to chat and on call tools like Slack, Telegram, email, Lark, PagerDuty, and similar services, with custom message templates I can edit. Use YAML config with environment variables, and keep the data locally so it works well for self hosting.
If you need to, look up the current docs online for anything unclear, but keep it practical and easy to run with Docker.
Are you gonna build this?
make sure you review the code using coderabbit