# senthilrch/kube-fledged

A kubernetes operator for creating and managing a cache of container images directly on the cluster worker nodes, so application pods start almost instantly

Repository: https://github.com/senthilrch/kube-fledged
Canonical: https://ross.abutalabs.com/products/kube-fledged
Language: Go
License: Apache-2.0
License Family: permissive
Topics: kubernetes, k8s, operator, image, cache, pull, delete, registry, repository, add-on, container, tool, golang, go
Last push: 2026-07-17T10:52:16+00:00

## Health v2 (maintenance only)
Score: 83/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 93, release rhythm 61, longevity 100
- inputs: {"age_days": 2868, "days_push": 47, "days_rel": 48, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1373, forks 127 (observed 2026-08-28T04:04:32.400476+00:00)

## What it is
kube-fledged is a Kubernetes operator that creates and manages a cache of container images directly on cluster worker nodes. By pre-pulling specified images onto chosen nodes, it enables application pods to start almost instantly without waiting for registry pulls.

## Use cases
- pre-pull container images onto kubernetes worker nodes
- speed up pod startup times in a kubernetes cluster
- reduce image pull latency when scaling applications rapidly
- cache images for real-time data processing workloads that need fast scale-out
- manage the lifecycle of a node-level image cache via CRUD APIs
- avoid registry rate limits and network delays during pod scheduling

## When to choose
- your kubernetes applications need near-instant pod startup or fast autoscaling
- you want to pre-pull images onto specific worker nodes declaratively
- image pull times from a remote registry are a bottleneck for your workloads
- you need a kubernetes-native add-on with Helm installation and CRUD APIs for image caching

## When to avoid
- your cluster already uses local registries or node image preloading via DaemonSets and that suffices
- you run a small cluster where image pull latency is not a concern
- you need general-purpose image building or registry mirroring rather than caching
- you require a fully managed cloud service rather than a self-managed operator

## Facets
- artifact type: application
- maturity: active
- function: container-orchestration, caching, deployment, developer-tools
- domain: cloud-computing, infrastructure-as-code, performance
- platform: go, cloud, self-hosted
- tags: kubernetes-operator, image-cache, pod-startup, container-images, helm-chart, cluster-addon, containers, devops, kubernetes, docker, linux

## Member repositories
- senthilrch/kube-fledged (main) score 83

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:32.400476+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:40:43.493235+00:00, confidence not recorded.
  - readme: https://github.com/senthilrch/kube-fledged (fetched 2026-08-28T04:04:32.400476+00:00, sha c6e7059eaa44)
- Data as of 2026-08-30T08:39:29.467469+00:00.