The self-managed Red Hat Enterprise Linux (RHEL) command-line assistant is now available as a Developer Preview. Building on the offline RHEL command-line assistant Developer Preview, this release shifts control to you: Deploy on your own infrastructure, and choose an LLM inference provider that aligns with your needs, whether disconnected or connected.
We have evolved the architecture to deliver a more flexible and powerful tool for RHEL system administrators. While we continue to offer our Red Hat-hosted RHEL command-line assistant experience for users who want a ready-to-use solution out of the box, the self-managed RHEL command-line assistant addresses the needs of organizations with offline/disconnected requirements, or the need to use specific LLM providers.
Here’s a look at what’s new and how you can get started.
New flexibility: Bring your own inference
The previous offline RHEL command-line assistant Developer Preview shipped with a bundled LLM and inference engine. In the new, self-managed Developer Preview, we have fundamentally shifted the architecture to give you more control.
The self-managed RHEL command-line assistant no longer relies on an included LLM model. Instead, it provides the flexibility to configure an LLM inference provider that aligns with your use case.
- For fully disconnected environments: Keep your data and queries on premise. For quick setups and testing, use the included template to connect the assistant to a RamaLama instance, or configure it to use Red Hat AI-based products for enterprise, at-scale inference.
- For connected environments: We include a template for Google Vertex AI. You can also configure several other connected LLM providers, allowing you to use the models that best fit your organization's AI strategy.
Smarter answers with the Red Hat Offline Knowledge Portal
The self-managed RHEL command-line assistant includes the Red Hat Offline Knowledge Portal, which houses a wealth of Red Hat expertise and knowledge.
When you ask the self-managed RHEL command-line assistant a question, it can dynamically retrieve information from the Offline Knowledge Portal and provide it as additional context to the LLM. This helps provide more accurate, RHEL-specific answers, even if your server has absolutely no external internet access.
Context-aware troubleshooting
General knowledge is helpful, but you also need context when diagnosing a system issue. You access the self-managed RHEL command-line assistant using the goose-redhatpackage, which is available in the RHEL extensions repository. From there, you can supercharge the assistant by enabling context-aware troubleshooting with Model Context Protocol (MCP) server for Red Hat Enterprise Linux.
By installing and using MCP server for Red Hat Enterprise Linux alongside the goose-redhat package, the assistant gains real-time access to your RHEL system’s actual state, including information from logs, performance metrics, and disk space utilization. This allows more targeted troubleshooting steps based on what’s actually happening on your system in real time.
Flexible, container-based deployment
To provide deployment flexibility, the self-managed RHEL command-line assistant is distributed as container images. This gives you the choice to run the containers directly on a RHEL system using Podman, or deploy them on Red Hat OpenShift.
How to get started
The self-managed RHEL command-line assistant is included as part of the Red Hat Satellite subscription. However, it operates independently, without relying on Satellite itself.
The self-managed RHEL command-line assistant represents a major step forward in bringing AI-powered operational guidance directly to where system administrators work the most: the command line. Learn more and get started by viewing the documentation.
