The command-line assistant

Review the most notable changes to the command-line assistant between RHEL 9 and RHEL 10.

Deprecation of the command-line-assistant package

The original command-line-assistant package, which is invoked by using the c command, is deprecated in RHEL 10. While the package remains available in the RHEL 9 and RHEL 10 AppStream repositories, no new features will be added, and the package will be removed in a future release. To maintain access to AI-assisted command-line tools and unlock new capabilities, transition your systems to the goose-redhat package. Transitioning to goose-redhat enables optional integration with the Model Context Protocol (MCP) server for RHEL Developer Preview.

For full migration instructions, see the Optimize Red Hat Enterprise Linux command-line assistant tasks using goose-redhat article.

Note

This deprecation applies only to the legacy command-line-assistant client. The underlying Red Hat-hosted command-line assistant backend infrastructure remains active and supported.
Direct RHEL interaction with the Goose CLI client
With this update, RHEL 10 introduces the Goose command-line interface (CLI) client (goose-redhat), a new AI-powered command-line assistant that can interact directly with the RHEL operating system when using it with the Model Context Protocol (MCP) server for RHEL Developer Preview. The client provides a standardized interface for integrating AI agents into your existing infrastructure workflows.
Disconnected and air-gapped deployment support for the command-line assistant

In RHEL 10, you can deploy the command-line assistant in disconnected environments. With this capability, you can use AI-assisted troubleshooting without requiring an active outbound connection to external public AI infrastructure.

Note

If you are an organization administrator planning RHEL 10 deployments in restricted networks, factor in the resource requirements for local Podman containers and model endpoint connections during system provisioning.
  • Container-based deployment

    In disconnected environments, you can deploy the assistant as a local container that you manage by using Podman.

  • Red Hat curated documentation

    Disconnected deployments integrate local Red Hat documentation sets within the container image to ensure accurate context and responses without internet access.

  • Third-party model integration

    You can configure the local container to route requests to approved on-premises or enterprise AI vendors and large language model (LLM) providers, such as OpenAI endpoints.

The command-line assistant for disconnected environments is available as a Developer Preview
You can run the command-line assistant in a completely disconnected, offline, or air-gapped environment, because it does not require any external network connectivity to operate locally on your workstation or on an individual RHEL system.
The command-line assistant supports image mode for RHEL
You can customize your Containerfile to include the command-line-assistant package, create a disk image from a container image, and boot a system with that image. The system image has the command-line assistant preinstalled, and you can use it after you register your system with subscription-manager.
The MCP server for Red Hat Enterprise Linux is available as a Developer Preview
You can enable AI applications to perform context-aware troubleshooting on RHEL systems, including log and performance analysis, by using the new Model Context Protocol (MCP) server for RHEL, available as a Developer Preview.
Color support for the command-line assistant
The command-line assistant supports color output by default, improving the output readability through increased visual contrast.
SAP Solutions documentation is available within the command-line assistant for disconnected environments as a Developer Preview
The Red Hat Enterprise Linux for SAP Solutions documentation is included in the command-line assistant knowledge base for disconnected environments as a Developer Preview. This inclusion provides more accurate and context-aware responses for SAP-related administrative and configuration tasks.