Menu
Triage RHOSO must-gather reports using OMC and the Core Operators Support Enablement guide
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
| name | support-triage |
| description | Triage RHOSO must-gather reports using OMC and the Core Operators Support Enablement guide |
| argument-hint | <must-gather-path> |
| user-invocable | true |
| allowed-tools | ["Bash","Read","Agent"] |
| context | fork |
You orchestrate RHOSO must-gather triage. You validate the environment, locate the must-gather, and dispatch the support-triage agent for diagnostics.
Check for OMC. Run which omc. If not found, tell the user:
OMC is required but not found in PATH. Install it with:
go install github.com/gmeghnag/omc@latestOr download a binary from https://github.com/gmeghnag/omc/releases
Then stop.
Locate the must-gather. The argument is a path to either:
.tar.xz archive (extract it first with tar xf)If no argument was provided, ask the user for the path.
Load into OMC. Run omc use <path>. The path should be the innermost must-gather directory (the one that contains cluster-scoped-resources/, namespaces/, etc.). If omc use is given a parent directory, it may not work. Look inside the provided path for the actual must-gather root:
find <path> -maxdepth 3 -name "cluster-scoped-resources" -type d
Use the parent of whichever directory is found.
Sanity check. Run omc get nodes. If it returns data, OMC is loaded. If it errors, report the problem and stop.
Detect namespaces. The default OpenStack namespaces are openstack (services) and openstack-operators (operators). Confirm they exist:
omc get ns openstack
omc get ns openstack-operators
If they have different names, ask the user or attempt to detect them from the must-gather directory structure (look under namespaces/).
Present these diagnostic categories to the user and ask which to run. The user may select one, several, or "all":
After setup and area selection, dispatch the support-triage agent:
Agent(
subagent_type="openstack-k8s-agent-tools:support-triage:support-triage",
description="Triage RHOSO must-gather",
prompt="Must-gather path: <path>
Namespaces: openstack=<detected-ns>, openstack-operators=<detected-ns-ops>
Selected categories: <user's selections>
Run diagnostics for the selected categories and produce a structured triage report."
)
If prior generic analysis findings were provided (when dispatched from /analyze-must-gather), include them in the prompt:
prompt="Must-gather path: <path>
Namespaces: openstack=<ns>, openstack-operators=<ns-ops>
Selected categories: <selections>
## Prior Generic Analysis Findings
<the generic analysis output>
RHOSO-specific resources detected. The generic analysis above identified surface-level
issues. Focus on RHOSO-specific diagnostics: correlate generic findings with RHOSO root
causes, check configurations the generic scan cannot evaluate (OperatorGroup
targetNamespace, NAD/NNCP alignment, StorageClass alignment, webhook namespace selectors),
and avoid duplicating already-reported issues unless adding RHOSO-specific context."
Present the agent's triage report to the user.
Scaffold RHOSO Deployment Topologies/Validated Architectures for the architecture repo with existing DT/VA analysis, kustomize generation, and staged automation
Interactive onboarding for openstack-k8s-operators — teaches Kubernetes operator fundamentals from scratch and tours the current operator repo at the learner's pace
Produce an analysis of observed problems in a must-gather report from OpenStack Services on OpenShift
Produce an analysis of observed problems in downloaded Zuul CI logs
Code review agent for openstack-k8s-operators following dev-docs conventions, lib-common patterns, and openstack-k8s-operators best practices
Validate CVE Jira tickets against the current codebase — assess reachability, exploitability, and recommend action (Go bump vs dependency bump vs code fix)