Supercharge RHEL troubleshooting with agentic AI: Introducing goose

AI is rapidly transforming how system administrators and developers interact with their environments. In 2025, we introduced the command line assistant, an optional tool available in Red Hat Enterprise Linux (RHEL) that allows users to ask questions and get natural language troubleshooting guidance right from the terminal.

We're now announcing the next step in our AI-driven systems management journey—the availability of the goose AI agent in the RHEL extensions repository for RHEL 9.8 and RHEL 10.2. A project of the Agentic AI Foundation, goose is a flexible, open source AI agent that can be used with the Model Context Protocol (MCP) server for RHEL (developer preview) to unlock new use cases for AI-assisted administration.

While the standard command line assistant provides a streamlined, read-only question and answer experience, gooseintroduces an extensible AI agent that helps enable context-aware system troubleshooting.

2 client options for accessing the Red Hat hosted command line assistant backend

When connecting to the Red Hat hosted command line assistant backend, administrators running RHEL now have 2 client pathways available. Depending on your requirements, you can choose between a traditional, streamlined Q&A tool or a newer, extensible tool with advanced features.

1. The goose client (goose-redhat)

For users looking to transition to a more powerful and capable assistant, we're introducing the goose-redhat package, available through the RHEL extensions repository on RHEL 9.8 and RHEL 10.2 systems. The goose-redhatpackage uses the open source goose AI agent to deliver advanced features such as support for MCP.  The goose-redhat package configures goose to communicate with the same Red Hat provided command line assistant backend.

2. The command line assistant client (command-line-assistant)

The original command-line-assistant package is available within the standard RHEL 9 and RHEL 10 AppStream repositories. Invoked using the c command, this client is designed for a low-overhead, read-only Q&A experience. It lets users ask natural language questions, pipe command outputs for context, and attach log files to receive quick troubleshooting advice and command suggestions.

The command-line-assistant package remains supported and available for use within the RHEL 9 and RHEL 10 systems. Please note, however,  that we are not planning on adding new features to this client in the future. 

Selecting the right package for your environment

To offer deployment flexibility, we're providing 2 distinct packages in the RHEL extensions repository.

1. The goose-redhat package: integrated and optimized

If you want an out-of-the-box agentic experience that is optimized for Red Hat environments, goose-redhat is your go-to.

Installing this package provides streamlined access to Red Hat's command line assistant backend infrastructure, which can incorporate information from the RHEL documentation and Red Hat Knowledgebase articles. Additionally, goose-redhat has strengthened security settings so it will prompt for manual approval before it calls tools.  

2. The goose package: Bring Your Own Model (BYOM)

For users who would like to provide their own inference infrastructure, the standalone goose package is available. This allows you to connect goose to external large language model (LLM) providers or local inference servers. This package provides a similar experience to the upstream goose and includes the default upstream settings.  

How to get started

The goose packages are distributed in the RHEL extensions repository, so you’ll first need to enable the repository on the RHEL system if you haven’t already.  

For RHEL 10:

sudo subscription-manager repos --enable=rhel-10-for-$(arch)-extensions-rpms

For RHEL 9:

sudo subscription-manager repos --enable=rhel-9-for-$(arch)-extensions-rpms

To install the goose-redhat package that is preconfigured with the Red Hat command line assistant backend:

sudo dnf install goose-redhat

Once installed, simply open a new shell session, run goose, and ask a question.

In this example, we ask if running RHEL 6-based containers on a RHEL 9 host is supported.  Goose uses the Red Hat command line assistant backend to answer the question, and provides a link to the source document on the customer portal.

$ goose
starting session | provider: rhel_cla model: RHEL-command-line-assistant
    session id: 20260527_2
    working directory: /home/brian
goose is running! Enter your instructions, or try asking what goose can do.
⚠️This tool uses AI technology. Do not include any personal information or other sensitive information in your input. Interactions may be used to improve Red Hat's products or services.
Context: oooooooooo 0% (0/128000 tokens)
( 0)> Is it supported to run RHEL 6 based containers on a RHEL 9 host?
Running RHEL 6 based containers on a RHEL 9 host is not supported.
Sources:
- [Red Hat Enterprise Linux Container Compatibility Matrix | Red Hat Customer Por
tal](https://access.redhat.com/support/policy/rhel-container-compatibility)
⏱️ Elapsed time: 17.16s
Always review AI-generated content prior to use.
Context: ●●oooooooo 15% (19660/128000 tokens)
( 0)> Press Enter to send, Ctrl-J for new line

Deep visibility with the Model Context Protocol (MCP)

One of the most exciting capabilities unlocked by goose is its support for the Model Context Protocol (MCP) server for RHEL which is available as a developer preview.

After installing and configuring MCP server for RHEL, goose can directly interact with either the local RHEL system it's installed on, or remote RHEL systems over SSH. This enables goose to provide context-aware advice and troubleshooting. Instead of manually copy-pasting log files, you can simply ask goose to investigate an issue, and goose will use MCP server for RHEL to review the relevant system information and logs. 

Refer to the instructions in the "Optimize the RHEL command-line assistant tasks by using goose-redhat" article for details about installing and configuring MCP server for RHEL for use with goose.  

Demo: Troubleshooting a filesystem

To show the power of goose combined with MCP server for RHEL, let’s look at a common real-world scenario—troubleshooting a nearly full root filesystem. Imagine receiving alerts that warn you that a RHEL system has reached 91% storage capacity on the root partition. You initiate a session with goose and provide a simple, natural language instruction: Help me figure out what's taking up so much disk space in the root filesystem.

The default configuration provided by the goose-redhatpackage prompts for permissions to call MCP tools, so we see a prompt asking for permission for goose to use the get_disk_usage tool provided by MCP server for RHEL.

Context: oooooooooo 0% (0/128000 tokens)
( 0)> Help me figure out what's taking up so much disk space in the root filesyste
m on this system?
─── get_disk_usage | rhel-mcp-server ─────────────────────────────────────
🔹 Goose would like to call the above tool, do you allow?
 │ ● Allow (Allow the tool call once)
 │ o Always Allow
 │ o Deny
 │ o Cancel
 └──

This tool returns information about filesystem utilization, and provides context to the LLM that the root filesystem is 91% full, with 23.5GB used.  Next, goose will prompt for permission to call the list_directories tool in the root of the filesystem.

The root filesystem `/` is at 91% utilization, with 23.5GB used. To investigate fu
rther, I need to find the largest directories within the root filesystem.
─── list_directories | rhel-mcp-server ─────────────────────────────────────
top_n: 10
order_by: size
path: /
sort: descending
🔹 Goose would like to call the above tool, do you allow?
 │ ● Allow (Allow the tool call once)
 │ o Always Allow
 │ o Deny
 │ o Cancel
 └──

This tool enables the LLM to see that the /home directory is using 17.1 GB of space.  Next, goose will prompt for permission to call the list_files tool on the /home directory to further narrow down what is taking up all that space.

The `/home` directory is consuming the most disk space on your root filesystem, with 17.1GB. The `/usr` directory is using 4.7GB, and `/var` is using 1.2GB.
Let's examine the `/home` directory further to identify the largest files within it.
─── list_files | rhel-mcp-server ─────────────────────────────────────
top_n: 10
path: /home
sort: descending
order_by: size
🔹 Goose would like to call the above tool, do you allow?
 │ ● Allow (Allow the tool call once)
 │ o Always Allow
 │ o Deny
 │ o Cancel
 └──

There are no files in the directory, so goose prompts for permission to call the list_directories tool on the /home directory.

There are no files directly in `/home` that are consuming significant space. It's likely that the disk space is being used by subdirectories within `/home`. Let's list the largest directories within `/home`.
─── list_directories | rhel-mcp-server ─────────────────────────────────────
top_n: 10
sort: descending
order_by: size
path: /home
🔹 Goose would like to call the above tool, do you allow?
 │ ● Allow (Allow the tool call once)
 │ o Always Allow
 │ o Deny
 │ o Cancel
 └──

Based on the information returned by MCP server for RHEL, the LLM can now see that the /home/brian directory is consuming 17.1 GB of space. Next, goose will prompt for permission to call the list_directories tool on the /home/brian directory. 

The `/home/brian` directory is consuming 17.1GB. Let's look inside `/home/brian` to find the largest directories.
─── list_directories | rhel-mcp-server ─────────────────────────────────────
order_by: size
top_n: 10
sort: descending
path: /home/brian
🔹 Goose would like to call the above tool, do you allow?
 │ ● Allow (Allow the tool call once)
 │ o Always Allow
 │ o Deny
 │ o Cancel
 └──

Goose is successfully able to narrow down that the /home/brian/virtual-machines directory is consuming 16.6 GB of space.

The `/home/brian/virtual-machines` directory is consuming 16.6GB of disk space. This is the largest contributor to the high disk usage on your root filesystem.
To free up space, you could consider the following:
* **Move or delete unused virtual machine images** within `/home/brian/virtual-machines`.
* **Investigate `/usr` (4.7GB) and `/var` (1.2GB)** for potential cleanup if clearing `/home/brian/virtual-machines` is not sufficient. Common areas in `/var` for large files include `/var/log` (logs), `/var/cache` (package manager cache), and `/var/tmp`. For `/usr`, it's usually system binaries and libraries, so direct cleanup might be limited unless there are unused packages.
* **Check `.local` (255.8MB) and `.cache` (240.7MB)** within `/home/brian`. These directories often contain user-specific application data and caches that can sometimes be cleared.
If you would like to investigate a specific directory further, please let me know.
⏱️ Elapsed time: 1m 57s
Always review AI-generated content prior to use.
Context: ●oooooooooo 8% (10842/128000 tokens)
( 0)> Press Enter to send, Ctrl-J for new line

Looking ahead

The evolution of systems management is moving rapidly toward AI-based tools. While the standard command line assistant remains available for read-only guidance, the introduction of goose provides a significant expansion of use cases for using AI to help with RHEL system management. Install goose on your RHEL system today and experience the next evolution of AI-assisted systems administration.