Stay in the know
Stay up to date on the latest Docker news, opinions and tools.
Welcome to the December edition of Docker Navigator. Missed an issue? Read past issues in our collection.
In this issue, we look at how developers are building smarter and more secure systems with Docker. You’ll see when a modular monolith can outperform microservices, how Docker Hardened Images and Docker Scout respond to real supply chain attacks, and how MCP and Sandboxes help you run safer AI agents. We also highlight faster model inference with vLLM, practical guidance for Hardened Images trials, and the latest on-demand sessions across AI and supply chain security.
You Want Microservices, But Do You Really Need Them?
Microservices can unlock flexibility, but they also add operational overhead that many teams do not need. With Docker, a modular monolith or SOA can deliver the consistency and scale teams expect while keeping governance, security, and development workflows manageable.
How Docker Hardened Images Patches Vulnerabilities in 24 hours
See how Docker Hardened Images patched Golang SSH CVEs in under 24 hours with Scout real-time CVE ingestion, automated rebuilds, SBOM clarity, and AI guardrails.
MCP Horror Stories: The WhatsApp Data Exfiltration Attack
How tool poisoning turned WhatsApp into a data exfiltration channel, and how Docker MCP Gateway blocked it with validation, network isolation, and audit logging.
Docker News
Introducing Docker Desktop 4.50: The Indispensable Development Platform
Docker Desktop 4.50 speeds up debugging with improved breakpoints, cleaner step-through behavior, and smoother IDE integration. The release strengthens enterprise security controls, improves Windows and WSL2 reliability, and introduces early MCP support so developers can start integrating AI into their workflows.
Securing the Software Supply Chain Shouldn’t Be Hard. According to theCUBE Research, Docker Makes It Simple
New research from theCUBE highlights how Docker improves supply chain security and reduces the effort required to stay compliant. Teams get clearer visibility, faster remediation, and guardrails that fit directly into existing workflows.
Why I joined Docker: Security at the center of the software supply chain
Mark Lechner, Docker’s CISO, shares his vision for a future where Docker not only powers the software supply chain, but actively safeguards it.
Security that moves fast: Docker’s response to Shai Hulud 2.0
Shai Hulud 2.0 created widespread supply chain risk, and Docker Scout responded by ingesting the malware indicators, scanning protected images, and correlating SBOMs to confirm exposure. The article highlights how this automated pipeline gave developers fast and reliable insight during an active attack.
Security that strengthens the ecosystem: Docker’s upstream approach to CVE-2025-12735
Docker identified the origin of the vulnerability in a nested open source dependency and partnered with the upstream project to ship a fix. The blog covers how this approach ensures the patch reaches all downstream users and reinforces the health of shared infrastructure.
Dive Deep: Build, Test, and Scale AI Agents with Docker
Explore how teams are configuring and connecting MCP agents, running secure agent workflows in Docker Sandboxes, accelerating model inference with vLLM, and evaluating image security more effectively with Docker Hardened Images.
Build and Secure AI Agents with Docker MCP
- Empower your agents to configure MCPs dynamically with smart search and code-mode tools in Docker MCP Gateway.
- Securely connect and manage remote MCP servers with built-in OAuth. Enable in two commands, then choose from 60+ servers in the Docker MCP Catalog and Toolkit.
- Discover the best MCP servers for developer productivity, including Context7, Obsidian, and Notion, and how to use them with MCP Toolkit.
Run Coding Agents Safely with Docker Sandboxes
Docker Sandboxes lets developers use coding agents without risking changes to the host machine. This post shows how container isolation, workspace mounting, and restricted permissions create a secure environment for agent-driven workflows.
Run High-Throughput Models with Docker Model Runner and vLLM
Docker Model Runner introduces support for vLLM to help teams run large models more efficiently on NVIDIA GPUs. The integration improves throughput for safetensors models and reduces configuration time with automatic engine routing.
Get More from Your Docker Hardened Images Trial
The first article in this new series helps you get the most out of your Docker Hardened Images trial. It covers the confusion points the docs do not address, the metrics that actually matter, and how to evaluate results with clarity.
Watch: AI Guide to the Galaxy
In the latest episode, Docker’s Oleg Šelajev meets with Jonas Scholz from E2B to break down how E2B sandboxes and Docker MCP enable secure execution environments for AI agents. The discussion covers sandbox architecture, controlled tool access, and how these systems protect developers while supporting complex agent workflows.
Around the Community
Connect with the Docker community to learn, ask questions, and stay up to date with what’s new in containerization. Explore the Docker Forum for active discussions and our Events page for what’s happening next. Here’s a recent thread from the forum:
- Docker 29 increased minimum API version, breaks Traefik reverse proxy
Traefik deployments that request an older Docker API version may fail after upgrading to Docker 29, which now requires API version 1.44 or higher. Updating Traefik or adjusting its Docker provider settings restores compatibility.
On the Calendar
This month offers a fully on-demand webinar lineup with sessions that help you improve AI workflows, strengthen your supply chain, and stay current with Docker’s latest tools. Explore the newest sessions:
- Building the Supply Chain for the Future, and How to Get There
Join Docker for a 7-part webinar series: Securing the modern software supply chain, and learn how to deliver secure, minimal, production-ready images with near-zero CVEs and up to 95% less attack surface. - Level up your local AI workflow with Model Runner
Learn how to run multi-modal models on virtually any GPU, streamline model distribution, and simplify collaboration across teams with Docker Model Runner.
That’s a Wrap
Thank you for reading Docker Navigator! If you enjoyed this issue, please share it with your friends so they can subscribe now.
Got any feedback or suggestions for our next issue? Send comments to newsletter@docker.com.