This write-up details CVE-2025-69902, a command-injection flaw in a kubectl MCP server that runs user input through a shell, letting an attacker execute arbitrary commands on the host.

More: https://ku.bz/2pxZf_bpJ

CPU and memory usage need context before you change Kubernetes requests and limits.

Throttling, pressure, and runtime events help explain the usage graph. Looking at each replica separately shows differences that a workload average can hide.

We are releasing the Kubernetes Resource Review Mixin, an open-source Grafana dashboard with Prometheus recording rules.

You can use it to:

- See how each Pod's CPU and memory usage compares to its requests and limits.
- Inspect CPU throttling, CPU and memory pressure when PSI metrics are available, and OOM and restart activity.
- Review replicas and workload revisions separately, with p95, maxima, and resource differences in the tables.

The mixin is an extension of our book on rightsizing Kubernetes workloads, The Technical Guide to Kubernetes Rightsizing.

Clone it here: https://github.com/learnk8s/resource-review-mixin

Hortator lets AI agents spawn sub-agents at runtime, with each agent running in its own pod with budget caps, network policies, PII redaction, and capability inheritance so children can never escalate beyond their parent's permissions.

More: https://ku.bz/KqYBl17Gv

This is a Hack The Box machine write-up that chains a LangFlow RCE and a forged JWT into a full Kubernetes cluster escape, step by step.

More: https://ku.bz/hw0Zs7JSb

LearnKube Day includes ten five-minute lightning talks from Kubernetes practitioners sharing practical lessons with the community.

Graziano Casto, Developer Advocate at Akuity, is one of the featured lightning-talk speakers at LearnKube Day Salt Lake City 2026.

The event brings together, lightning talks, YAML Games, and a community reception with a live DJ(our own Amin Astaneh).

Learn more: https://learnkube.com/learnkube-day-salt-lake-city-2026

This week on Learn Kubernetes Weekly 206:

🌐 DNS Is Weird Inside Kubernetes on AWS
⏳ TIL: Kubernetes ConfigMap Volume Mounts Are Eventually Consistent — Not Immediately Consistent
🚨 CVE-2025-69902: OS Command Injection in kubectl-mcp-tool
🐘 Building the Foundation for PostgreSQL Lifecycle Management in a DBaaS on Harvester
💰 We Cut Our Cassandra Bill From $30K to $10K a Month by Self-Hosting on Kubernetes

Read it now: https://kube.today/issues/206

⭐️ This issue is brought to you by Teleport — stop sharing credentials and give every engineer, service, and AI agent a login that expires automatically https://ku.bz/2tJxlTyDy

This article explains why secrets and tokens form the access system of a platform rather than plain config, and how static secrets and short-lived tokens together decide who can reach what.

More: https://ku.bz/YTvRcJgW7

KubeClaw is a production-grade Helm chart for running OpenClaw AI agents on Kubernetes with secure defaults, pinned images, default-deny egress via Blocky, digest-locked upgrades, and batteries-included observability.

More: https://ku.bz/CbKpPNvJq

This article explains how GKE Agent Sandbox uses gVisor to isolate untrusted AI-generated code with separate node pools, Workload Identity, RBAC, and security observability.

More: https://ku.bz/0dy9y1wPw

This tutorial teaches how to migrate EKS from AWS VPC CNI and kube-proxy to Cilium using blue/green nodes, WireGuard encryption, and Hubble for observability.

More: https://ku.bz/sYGLb0l37

Seven provisioning models. One central Kubernetes team.

Molly Sheets joins Daniele and Bart to discuss the staffing cost of supporting multiple Kubernetes tenancy models.

- Assigning ownership across tenancy models
- Giving AI Ops cross-team context
- Planning canary and cell-based releases

Watch: https://ku.bz/Db3VnC3nl

This episode is sponsored by LearnKube. When a release breaks production, can your developers and platform engineers troubleshoot it together? Request private Kubernetes training tailored to your platform and workloads, with live instruction and 60% hands-on labs.

Even as Kubernetes platforms mature, securing them does not necessarily become easier.

Marina Moore from Edera explains that as Kubernetes becomes more complex, both cybersecurity and explainability become more difficult. The main issue is that adding more components means more systems to understand and more areas to protect.

Watch the full interview: https://ku.bz/pzn5hfmTf

This week on Learn Kubernetes Weekly 207:

🚦 Load Testing Apollo Router
⚖️ Client-Side Load Balancing at a Million Requests per Second
🦞 KubeClaw: OpenClaw for Adults
🚀 We Scaled an Insurance Batch Job 4000x — Here’s Every Bottleneck We Hit
👁️ How I Built a Production-Grade Object Detection System That Scales Itself

Read it now: https://kube.today/issues/207

⭐️ This newsletter is brought to you by LearnKube — understand how Kubernetes works, and what to do when it breaks. Live training with 60% hands-on labs https://ku.bz/hypSbyc-V