You're pledging to donate if the project hits its minimum goal and gets approved. If not, your funds will be returned.
$1,500 left to reach the funding goal.
Project summary:
Let's be real, AI agents managing servers is a terrifying idea right now. If an LLM hallucinates a command, a company could go completely dark. I built AIOps Cockpit to fix that. The AI can look at server logs and write a code fix, but it's physically blocked from executing it. It can only open a Git Pull Request. A real human has to read the code and click merge. It's live right now and fully open source.
What are this project's goals? How will you achieve them?
For the next 3 months, I'm just going to lock myself in and make this safety layer bulletproof. I need to write out a real threat model for prompt injections. I also want to hire a security guy to try and break my Git preflight checks (basically making sure the AI can't sneak a bad command through). And I'll probably add Kubernetes support because that's what everyone uses now.
How will you know it worked?
It worked if the independent security audit passes without any major holes, and if I can get at least 3 real DevOps teams to actually use it on their own servers.
How will this funding be used?
Honestly, it's mostly to buy me time. 1500 bucks covers about 3 months of my rent and food so I don't have to take freelance gigs. The rest goes to Claude/Grok API tokens so I can run thousands of fake attacks against my own AI to test the defenses.
Who is on your team? What's your track record on similar projects?
It's just me. I'm a 20-year-old student currently doing my B.E. in Electronics & Telecom. I already built and shipped the whole thing by myself (React, Node, Socket.IO). The code is on my GitHub. I've also applied to SINE IIT Bombay and AWS Activate for support.
What are the most likely causes and outcomes if this project fails?
The biggest risk is that companies just aren't ready to let AI anywhere near their servers yet, even with the guardrails. If that happens, I won't really care, because the code is open source. Other researchers can just use my Git-based approach to build their own safer agents.
How much money have you raised in the last 12 months, and from where?
Nothing. Zero. I'm completely bootstrapped.