The three things a small k3s cluster needs after "it runs": knowing what it's doing, somewhere to put files and sites, and backups that have actually come back at least once. Three tested Flux recipes, one price. Made for people who run k3s at home or in a closet (the cluster, not you) and have an hour a week for ops. What's in it Versions, each the product's current download: monitoring 1.1.0, S3 + static sites 1.2.0, backups 1.2.0. Monitor it. Prometheus, Grafana, Loki, probes, a Golden Signals dashboard and email alerts, in about 1.5 GiB of RAM. On three nodes, optionally highly available. Host on it. Your own S3 and static sites with the setup as code, public through Cloudflare Tunnel with zero open ports. Back it up. MariaDB, Postgres to the minute and S3, with weekly restore tests that alert you when they fail. Why the bundle All three come from the same running home lab, mine, so they were made to fit together: the S3 recipe creates the bucket the backups recipe needs, and the monitoring recipe adds dashboards and alerts for both. The bundle costs €13 less than buying them one by one, which is roughly one very nice lunch in Berlin. Made for AI coding agents Each recipe ships with an AGENTS.md and a CLAUDE.md written as an operating procedure. Your agent asks you before it changes anything. How it was tested ✅ Each recipe deployed from an empty VM, following its README ✅ Monitoring with S3, and S3 with backups, tested together in one cluster ✅ Each recipe tested on x86_64 and again on arm64 All three at once in a single cluster wasn't tested, because that needs a bigger machine than the one I used, so I won't claim it. Is it for you? Yes, if you run k3s on one to a few small nodes and want monitoring, storage and backups you can trust, all in Git. No, if you need high availability on one or two nodes, or run 32-bit ARM. With three nodes, monitoring and S3 each have a tested option to survive losing one; the example Postgres in the backups recipe still runs once.