kubeaid-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| prompts | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_contextsA | List the kubeconfig contexts (clusters) this server can target, including which one is the default and which are write-protected. |
| list_namespacesA | List namespaces in the cluster with their status and age. |
| list_nodesA | List cluster nodes with Ready status, roles, kubelet version and internal IP. |
| get_eventsA | Recent cluster events, newest last. Filter to one namespace and/or Warning-type events only. |
| list_argocd_appsA | List the ArgoCD Applications KubeAid manages in this cluster, with sync status (Synced/OutOfSync), health (Healthy/Degraded/...), deployed revision and whether auto-sync is on. KubeAid disables auto-sync by default, so OutOfSync apps are usually awaiting a deliberate sync. Start here for any GitOps question. |
| describe_argocd_appA | Full detail for one ArgoCD Application: Git/Helm sources, sync and health status, error conditions, the resources that are OutOfSync or unhealthy, last sync operation result and recent deploy history. Use this to diagnose why an app is Degraded or a sync failed. |
| list_podsA | List pods with derived status (CrashLoopBackOff, ImagePullBackOff, Terminating, ...), ready count, restarts, node and age. |
| describe_podA | Deep-dive one pod: per-container state (waiting/termination reasons, last crash exit code), conditions and the pod's recent events. |
| get_pod_logsA | Tail a container's logs. Set previous=true to read the prior (crashed) instance's logs — essential for CrashLoopBackOff. |
| list_deploymentsA | List deployments with ready/up-to-date/available replica counts and age. |
| describe_resourceA | Fetch any Kubernetes resource (including CRDs like SealedSecret or Certificate) by apiVersion, kind and name. Returns the full object. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| diagnose_app | Figure out why a KubeAid-managed ArgoCD app is Degraded, OutOfSync or failing to sync. |
| sync_review | Review every OutOfSync app before syncing — KubeAid keeps auto-sync off so drift needs a human decision. |
| cluster_health_check | Full health sweep of a KubeAid cluster: nodes, ArgoCD apps, workloads and warning events. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 11 tools
Each tool targets a distinct resource/action - nodes, events, ArgoCD apps, pods, deployments, generic resource fetch, contexts, namespaces - with no overlapping purposes. The list/describe pairs for pods and ArgoCD apps are clearly separated by summary vs. detail, and describe_resource is explicitly generic.
Naming follows a consistent verb_noun pattern: list_* for summary listings, describe_* for detailed views, and get_pod_logs as a specific action. Even though get_events uses 'get' instead of 'list', the pattern is predictable and readable, so deviation is minor and not confusing.
With 11 tools, the set is well-scoped for a Kubernetes diagnostic server. Each tool addresses a common troubleshooting need (node health, pod issues, logs, ArgoCD sync, deployments) without excess or missing core operations.
The surface covers the main diagnostic workflows: listing and describing nodes, pods, deployments, and ArgoCD apps; retrieving events and logs; fetching any resource via describe_resource; and listing contexts/namespaces. For a read-only troubleshooting tool, this is comprehensive with no critical dead ends.