Skip to main content
Glama
akkireddy-challa

k8s-mcp-server

by akkireddy-challa
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault

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

CapabilityDetails
tools
{
  "listChanged": false
}
prompts
{
  "listChanged": false
}
resources
{
  "subscribe": false,
  "listChanged": false
}
experimental
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
list_namespacesA

List all namespaces present in the connected Kubernetes cluster.

Usage Guidelines

  • Call this tool first when discovering available cluster domains or when the user does not supply an explicit namespace.

  • Do not use this tool to inspect workload health; use list_pods instead.

Behavioral Transparency

  • Read-only operation.

  • Scans cluster-wide namespace resources using CoreV1Api.

  • Returns an alphabetically sorted list of namespace names with total count.

list_podsA

List pods within a specified namespace along with their lifecycle phase and IP address.

Usage Guidelines

  • Use this tool when identifying running workloads, crash loops, or verifying pod readiness.

  • If troubleshooting a failing pod, pass its name to get_pod_logs or call get_cluster_events.

Behavioral Transparency

  • Read-only operation.

  • Omits internal container state arrays; highlights high-level phase (Running, Pending, Failed).

get_pod_logsA

Extract stdout and stderr log streams from a pod container for diagnostics.

Usage Guidelines

  • Use when investigating application crashes, HTTP 500 errors, or startup exceptions.

  • Keep tail_lines small (e.g. 50–200) to avoid overloading LLM token context.

Behavioral Transparency

  • Read-only query.

  • Returns raw text log lines or a descriptive error if the pod or container does not exist.

get_cluster_eventsA

Retrieve Kubernetes cluster events to diagnose scheduling failures, node pressure, or eviction warnings.

Usage Guidelines

  • Call this tool when pods remain stuck in 'Pending' or 'CrashLoopBackOff' states.

  • Use warning_only=True to eliminate normal informational noise during triage.

Behavioral Transparency

  • Read-only query.

  • Returns up to the 50 most recent events formatted with event type, target resource, reason, and message.

list_deploymentsA

List deployments in a namespace with desired vs available replica counts.

Usage Guidelines

  • Use this tool to verify rollout status and check whether workloads meet desired scale.

  • When replicas show 0 available, use list_pods and get_cluster_events to locate the failure.

Behavioral Transparency

  • Read-only operation using AppsV1Api.

  • Returns name, desired replicas, and ready replica count.

list_servicesA

List Kubernetes services, exposed ports, and service types within a namespace.

Usage Guidelines

  • Use this tool when checking network ingress, internal DNS resolution, or service endpoints.

  • Complement with list_pods to confirm underlying backend endpoints exist.

Behavioral Transparency

  • Read-only query using CoreV1Api.

  • Lists service name, type (ClusterIP, NodePort, LoadBalancer), and mapped ports.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A4.7/5.0

Scored across 6 tools

Disambiguation5/5

Each tool targets a distinct Kubernetes resource or action: namespaces, pods, logs, events, deployments, and services. No functional overlap that would confuse tool selection.

Naming Consistency4/5

Names follow a consistent lowercase snake_case verb_noun pattern, but four use 'list_' while two use 'get_'. The split is semantically reasonable yet not perfectly uniform.

Tool Count5/5

Six tools form a focused, readable set for read-only Kubernetes inspection and troubleshooting, without redundant or excessive entries.

Completeness5/5

The set covers the core resources needed for workload troubleshooting: namespaces, pods, logs, cluster events, deployments, and services. There are no obvious dead ends for the inferred read-only diagnostic domain.

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources