ml-stack Helm chart
quench-ml-stackVerified publisherScored 27 Sept 2026
Hardened machine-learning stack: JupyterHub notebooks, the MLflow tracking server and model registry, and Label Studio for data labeling, wired together in one install. All QuenchWorks images are nonroot, 0-CVE, pinned by digest and cosign-signed.
Latest 0.0.2app version 1.0.0 (not verified against the render) 0Artifact Hub
ml-stack 0.0.2 deploys 5 container images: ghcr.io/quenchworks/images/jupyterhub, ghcr.io/quenchworks/images/configurable-http-proxy, ghcr.io/quenchworks/images/label-studio, ghcr.io/quenchworks/images/mlflow and 1 more. Across them, 56 findings — 0 critical, 0 high. The highest contribution is CGA-2mrq-cv8h-r8wg in grpc-1.81 1.81.1-r6, with no fix listed.
Radar Score
Radar Score is the sum of every finding’s contribution; the colour is the worst finding’s band.
Container images
| Image | Tag | Vulnerabilities | Radar Score |
|---|---|---|---|
| ghcr.io/ | digest-pinned | 00010 | 62 |
| ghcr.io/ | digest-pinned | 0006 | 32 |
| ghcr.io/ | digest-pinned | 00212 | 114 |
| ghcr.io/ | digest-pinned | 00015 | 105 |
| ghcr.io/ | digest-pinned | 00010 | 62 |
Rendered with the chart’s default values for linux/amd64. Optional subcharts, images set by operator flags and images inside CRD payloads are not seen and are counted as unmeasured.
Medium findings
Medium: findings whose contribution to the Radar Score is 15–39. Show every band
| Severity | Advisory | Package | Fixed in |
|---|---|---|---|
| Medium | CGA-2mrq-cv8h-r8wg | grpc-1.81 | no fix listed |
| Medium | CGA-qvpw-v4vp-864h | grpc-1.81 | no fix listed |
Indexed versions
The latest version and the previous major, as selected nightly from the repository’s index.
README badge
Markdown. The badge shows the latest version’s band and Radar Score, refreshed daily.