{"author":"prateekdalal","children":[{"author":"prateekdalal","children":[],"created_at":"2026-02-28T10:03:52.000Z","created_at_i":1772273032,"id":47193087,"options":[],"parent_id":47193086,"points":null,"story_id":47193086,"text":"Over the last 48 hours we shipped a major update to our open-source observability + infra + AI runtime stack.<p>Big changes:<p>Renamed kakveda-aankh \u2192 kakveda-netra<p>Built as a real host-level agent (CLI + systemd support)<p>Background mode (--start, --stop, --bg-status)<p>Full infra metrics:<p>CPU, memory, disk, network, load<p>Docker container metrics<p>temperature, FD usage, diagnostics<p>Kubernetes inventory:<p>nodes, pods, deployments, services, configmaps, secrets<p>kubeconfig auto-detection<p>Dashboard upgrades:<p>golden signals<p>trend charts<p>thresholds + alert coloring<p>detailed drilldowns<p>Custom dashboards:<p>saved layouts<p>YAML import&#x2F;export<p>Runtime control of host agent from dashboard<p>Cost comparison docs (Datadog, AppDynamics, Logz.io, LangSmith etc.)<p>The idea is simple:\nInfra monitoring + AI&#x2F;agent governance + observability in one platform.<p>Still early. Would love feedback from infra engineers and K8s folks.\nGithub : <a href=\"https:&#x2F;&#x2F;github.com&#x2F;prateekdevisingh&#x2F;kakveda\" rel=\"nofollow\">https:&#x2F;&#x2F;github.com&#x2F;prateekdevisingh&#x2F;kakveda</a>","title":null,"type":"comment","url":null}],"created_at":"2026-02-28T10:03:52.000Z","created_at_i":1772273032,"id":47193086,"options":[],"parent_id":null,"points":1,"story_id":47193086,"text":null,"title":"Kakveda open source-level AI, Infra observability agent","type":"story","url":"https://www.kakveda.com/"}
