Show HN: Protecting Infra from Humans and Agents

1 phrocker 1 7/30/2025, 3:09:58 PM github.com ↗

Comments (1)

phrocker · 11h ago
A few years ago, I started on a project and recently open sourced / sanitized part of it. RDP and memory store is still being sanitized (removing names, etc).

I began using the project because I wanted to learn about AI but also have workers (which I now call agents) use rules to perform actions on my behalf and monitor what I'm doing.

In short, I did this as a learning exercise. I just open sourced it and haven't really passed it along to anyone but am posting here in case anyone wants a view into the history of what people have been doing in AI. I did this with a small group before MCP, before semantic kernel, etc.

I wanted a single plane of glass to access infra, and use agents to do things like automate tasks, etc.

Chose java because that is what the group knew best. I don't expect anyone to want to use it, but I love to tinker and it's something I've enjoyed working on since I don't normally do UI work.