Product upvotes vs the next 3

Waiting for data. Loading

Product comments vs the next 3

Waiting for data. Loading

Product upvote speed vs the next 3

Waiting for data. Loading

Product upvotes and comments

Waiting for data. Loading

Product vs the next 3

Loading

Stakpak 3.0 CLI

Open source DevOps agent for devs who just want to ship

Stakpak is a fully open source DevOps agent written in Rust that helps developers secure, deploy, and operate production infrastructure from the terminal or in GitHub Actions You can run it locally, bring your own keys, or use it with self-hosted models, while keeping safety built in from day one. Stakpak is designed to work reliably with real production infrastructure. Try it now: curl -sSL https://stakpak.dev/install.sh | sh

Top comment

Thanks for the support on our earlier launch.

We’re launching Stakpak again, now fully open-sourced under Apache 2.0, and it’s been one of the trending Rust projects on GitHub over the past week.

Most LLMs are not built for DevOps. They leak credentials, don’t understand real infrastructure, and break down on non-trivial upgrades or deployments. You either babysit every command or keep them far away from production.

Stakpak is our attempt to solve this properly. It’s an open-source DevOps agent designed with safety and determinism as first-class constraints. The agent runs through a controlled execution layer, uses MCP over mTLS, supports async tool calls, and streams progress in real time for long-running operations. Secrets are dynamically substituted so the model never sees them.

Rulebooks are a core part of the system. They’re explicit guidelines that teach the agent how to handle real DevOps tasks correctly, reducing time and complexity without relying on fragile prompts or hidden behavior.

You can run Stakpak locally in your terminal, inside GitHub Actions, or trigger it via Slack using our SaaS. Everything is open, including system prompts.

Our long-term goal is to make infrastructure increasingly self-driving. This is just the beginning.

Would love feedback from engineers who’ve actually operated systems in production. Try it out, star us on GitHub if it’s useful, and follow along for updates.