@notharshhaa/devops-mcp

devops-infra MCP Server

Unified MCP server for DevOps engineers — query and manage Kubernetes, ArgoCD, Prometheus, and PagerDuty from any MCP-compatible AI agent.

VerifiedInstall Ready
devops-infradevops-infrakubernetes
4 views1 stars1 forksv2.1.0MIT

Why This Matters

Discovered via github-topic:model-context-protocol and last synced 2mo ago.

VerifiedInstall Ready
Source
github-topic:model-context-protocol
Stars
1
Last synced
2mo ago
Install
Instructions detected

Install

1. Install the package

npx @notharshhaa/devops-mcp

2. Add to claude_desktop_config.json

{
  "mcpServers": {
    "-notharshhaa-devops-mcp": {
      "command": "npx",
      "args": [
        "@notharshhaa/devops-mcp"
      ]
    }
  }
}

Config file location: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) / %APPDATA%\Claude\claude_desktop_config.json (Windows)

73
Tools
0
Resources
0
Prompts
Standard I/O
Transport

Available Tools (73)

Prefix

Provider

k8s__list_pods

read

k8s__delete_resource

destructive

argo__get_app_diff

read

prom__query

read

prom__list_targets

read

pd__add_note

mutate

k8s__describe_resource

read

argo__sync_app

mutate

prom__summarize_service_health

read

pd__who_is_oncall

read

Prometheus

HTTP API (PromQL)

k8s__get_events

read

k8s__switch_context

mutate

argo__get_resource_tree

read

prom__query_range

read

prom__metric_metadata

read

pd__get_incident

read

pd__summarize_incident

read

PagerDuty

REST API v2

k8s__list_deployments

read

Loki

HTTP API (LogQL)

k8s__apply_manifest

mutate

argo__terminate_op

mutate

logs__get_recent_errors

read

pd__get_log_entries

read

k8s__get_resource_usage

read

k8s__rollout_restart

mutate

argo__get_app

read

prom__get_firing_alerts

read

k8s__get_pod_logs

read

k8s__list_contexts

read

argo__get_app_history

read

prom__label_values

read

pd__list_incidents

read

pd__escalate_incident

destructive

k8s__scale_deployment

mutate

argo__list_apps

read

argo__rollback_app

mutate

prom__list_alerts

read

Tool

Tier

pd__list_services

read

devops__debug_service

read

logs__search

read

pd__acknowledge_incident

mutate

k8s__get_node_status

read

k8s__list_cronjobs

read

prom__slo_status

read

helm__list_releases

read

helm__get_status

read

helm__rollback

mutate

devops__incident_timeline

read

Symptom

What it checks

oom

OOMKilled events → memory usage → pod describe → resource limits

Helm

CLI (helm binary)

k8s__get_network_policies

read

helm__get_values

read

crashloop

Pod status → logs (tail 50) → BackOff events → deployment health

k8s__get_hpa

read

devops__explain_change

read

k8s__diff_resource

read

pod-pending

Scheduling events → pending pods → node capacity

k8s__get_ingresses

read

prom__compare_periods

read

helm__get_history

read

high-latency

p95 latency → resource usage → firing alerts → recent deploys

k8s__get_cronjob_status

read

k8s__list_pvcs

read

devops__runbook

read

5xx

Error rate → Loki error logs → deployment health

ArgoCD

REST API

k8s__list_services

read

devops__health_report

read