Empower your AI tools with new agent skills for Red Hat Enterprise Linux

AI tools are transforming how system administrators and developers manage their infrastructure,  but when using generic AI assistants to troubleshoot Red Hat Enterprise Linux (RHEL) systems, the advice can sometimes lack distribution-specific context. An AI tool might assume a different Linux environment, suggest commands from other distributions that do not apply to RHEL, or recommend disabling critical security controls to resolve a permissions issue.

To help bridge this gap, we are introducing 2 new integrations, currently in developer preview, designed to bring Red Hat knowledge directly into your AI tools: the translator agent skill for RHEL and the best practices agent skill for RHEL.

Built on the open Agent Skills standard , these integrations equip supported AI agents with the vocabulary, workflows, and domain expertise to help manage RHEL systems more effectively.

Translator agent skill for RHEL

The translator skill for RHEL enables AI tools to translate general Linux concepts, commands, and terminology into RHEL equivalents, so AI-generated advice aligns with RHEL standards. 

The translator skill enables AI tools to move beyond generic advice, and instead provide the RHEL equivalent for tasks while educating you and your team on the technical rationale.

Key capabilities:

• Package management: Translating apt, dpkg to dnf, rpm, and RHEL Application Streams.
• Containers: Mapping docker, docker-compose, and Dockerfiles to RHEL's daemonless podman, Quadlet, and Containerfiles.
• Networking and firewalls: Converting ifconfig, netplan, ufw, and iptables to nmcli, nmstatectl, and firewalld.
• Security and compliance: Guiding users from AppArmor and Tripwire to SELinux, AIDE, and OpenSCAP.
• Service and log management: Redirecting legacy init scripts or /var/log/syslog queries to systemd, journalctl, and /var/log/messages.
• Upgrades and migrations: Translating dist-upgrade or cross-distribution migration requests to native tools like Leappand Convert2RHEL.

Best practices agent skill for RHEL

When troubleshooting complex systems, administrators need solutions that align with RHEL best practices, rather than generic or outdated advice found on public forums. The best practices skill empowers AI tools with Red Hat’s recommended frameworks for diagnostics, health monitoring, and proactive maintenance.

By following structured diagnostic workflows, your AI assistant can provide tailored guidance grounded in Red Hat knowledge.

Key capabilities:

• SELinux: Assisting with diagnosing "Permission denied" errors and Access Vector Cache (AVC) denials.
• Kernel operations: kdump configuration, crash analysis (vmcore), and kernel live patching (kpatch).
• System performance: Proactive health checks and Performance Co-Pilot (PCP) monitoring.
• Security and compliance: Managing cryptographic policies, FIPS mode, and compliance scanning with OpenSCAP.
• Support readiness: Generating sos reports for Red Hat Support.
• Subscription management: Helping resolve common subscription management and repository availability issues.

Context-aware diagnostics with the MCP server for RHEL

While the skills provide the domain knowledge, AI assistants become significantly more effective when they understand the specific environment they are analyzing. By pairing these skills with the Model Context Protocol (MCP) server for RHEL (also available as a developer preview), your AI tool can also directly interact with your RHEL hosts.  

This integration allows the AI to determine the exact version of RHEL in use, check service statuses, or read specific log entries before formulating a response. By grounding AI-generated advice in the actual, real-time state of your system, the recommendations become more contextual, accurate, and actionable.

See the translator skill in action

I’m using goose and a locally hosted open source large language model (LLM). I haven’t installed the translator skill yet, and I’ll ask the following question within goose: I'm migrating to RHEL; is btrfs something that is available in RHEL 10?

Without the translator skill, the LLM responds with information that is totally inaccurate:  

( 0)> I'm migrating to RHEL; is btrfs something that is available in RHEL 10?
Yes, **btrfs is available in RHEL 10**. It has been supported in RHEL for several releases now:
- **RHEL 8**: btrfs was available as a supported filesystem option (though xfs was the default)
- **RHEL 9**: Continued btrfs support
- **RHEL 10**: btrfs remains a supported filesystem

All of this information is completely wrong—btrfs was never available or supported in RHEL 8, 9, or 10.  

Next, I’ll make the translator skill available within goose, and ask the same question, with the same AI model: I'm migrating to RHEL; is btrfs something that is available in RHEL 10?

The AI model identifies that the translator skill could help with this question, and goose prompts me for permission to load the skill:

── loadSkill | skills ──────────────────────────────────
name: rhel-translator
◆  Goose would like to call the above tool, do you allow?
│  ● Allow (Allow the tool call once)
│  ○ Always Allow
│  ○ Deny
│  ○ Cancel
└─

Using knowledge that it gained from the translator skill, the LLM correctly responds that btrfs is not supported or included in RHEL 10, and that XFS and Stratis are other options available in RHEL: 

On RHEL, **btrfs is not supported**. Red Hat does not include btrfs kernel support or packaging in RHEL 10.
The RHEL-native equivalents are:
- **XFS** – the default filesystem on RHEL, used for all standard installations. It's a high-performance, scalable filesystem with features like online defragmentation and growing.
- **Stratis** – if you want pool-based management with snapshots, thin provisioning, and self-healing (features often associated with btrfs or ZFS), Stratis is RHEL's modern answer. It sits on top of LVM and provides simplified volume management.

Get started today

Both the translator and best practices skills for RHEL are currently available as a developer preview. Because they use the open Agent Skill standard, they're designed to be client-agnostic and can be integrated into modern AI agents or development environments that support the format.

To try them out:

1. Download the SKILL.md files for the translator and best practices skills.
2. Load them into your preferred AI agent (goose, Cursor, Claude Code, etc.), consulting your specific tool's documentation for import instructions.
3. Configure the MCP server for RHEL in your client to enable context-aware, real-time system diagnostics.

These skills are built to plug into the tools you already use. Download them today to bring Red Hat expertise into your preferred AI tool. Visit the Red Hat Ecosystem catalog to learn more and stay up to date on the agent skills, agents, and MCP servers available across the Red Hat product portfolio.