Screenshot of reddit post: Recently ran into a problem that I figure isn’t unique to me. I got into self hosting when I made a personal agent that proactively keeps me on track when I work. I hosted it on tailscale. Then I added LLM chat. And then coop games I made to play with my friends. Been a blast. But to make it useful I put a lot of data and access in there I don’t really want others to access like realtime laptop/phone screenshots, GPS coordinates, email etc. Usually this is fine because saas will handle security for me but now I’m responsible. And tailscale and aws/gcp/azure don’t have centralized security alerts. Basically I wanted a security camera for my network/apps so I know what’s happening in my systems in case anyone tried to break in. Logs help a bit obviously, but there's too much and having an agent read them constantly sounds really expensive. I just want to answer: who’s trying to attack me and potentially mucking around inside? I ended up building a security tool that runs inside my network and cloud and watches the network, VMs and auth logs, then sends alerts to me when something is off. I know there’s tools like wazuh or elastic but they seem to be built for security professionals. I have to set up a bunch of rules and alerts which is hard and it’s a time suck to figure out what to include. I tried using codex and claude but it just makes dumb security decisions I can’t trust at all. The tool I made scans the network and VMs to configure rules and alerts automatically and deterministically without AI at all. If you guys have any good ideas here, lmk.