Read-only MCP server for querying OpenShift Compliance Operator resources via Kubernetes API
All 8 tools lack input parameter descriptions entirely. While tool names follow verb_noun convention (list_*), descriptions are minimal (20-60 chars, well below the 194-char baseline). No input schemas are visible in the source code, all tools declare empty input {} with no type information. Output is unstructured JSON strings rather than typed objects. No error handling, validation, or recovery guidance. No pagination despite listing potentially large result sets. This server fails basic definition quality standards.
List all ComplianceCheckResult objects in 'openshift-compliance'. Returns JSON array as a string. NOTE: If you want to filter by scanName, you'd do label selectors in a more advanced approach.
List all ComplianceRemediation objects in 'openshift-compliance'. Returns JSON array as a string. This is read-only, so no patching/applying.
List all ComplianceScan objects in the 'openshift-compliance' namespace. Returns JSON array of items as a string.
List all ComplianceSuite objects in the 'openshift-compliance' namespace. Returns JSON array of items as a string.
List all ProfileBundle objects in 'openshift-compliance'.
List all Profile objects in 'openshift-compliance'.
No input schemas defined.
Descriptions are minimal (15 - 35 chars) and lack context. None explain WHEN to call the tool, what structure is returned, or how to use results. Baseline is 194 chars; these are 10 - 20% of that.
No pagination support. Tools return raw JSON arrays with no limit, offset, or cursor. Large result sets will blow context windows and degrade LLM reasoning.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 37 | 2026-07-28+ | v2 |
List all ScanSetting objects (plural: 'compliancescansettings') in the 'openshift-compliance' namespace. Returns JSON array of items as a string.
List all TailoredProfile objects in 'openshift-compliance'.
Output is unstructured JSON strings (json.dumps). No typed response schema. LLMs must parse raw JSON, wasting tokens and inviting parsing errors.
No error handling or recovery guidance. If a Kubernetes API call fails, the exception propagates with no actionable message for the LLM.