convez.ai
MCP ServerDevOps & InfraestruturaAvançado
arnstarn

MCP Server Spotinst

arnstarn/mcp-server-spotinst

Gerencie infraestrutura Spotinst (AWS/Azure) via Claude com 34 ferramentas

Ver no GitHub
InicialPluginScore · de 0 a 100

Servidor MCP que integra a API Spot.io (Spotinst) ao Claude, permitindo gerenciar clusters Ocean, Virtual Node Groups, Elastigroups e analisar custos em AWS e Azure. Suporta operações de leitura e escrita com acesso multi-conta, além de análise de right-sizing e trending de custos.

Instalação

$npx convez-ai arnstarn/mcp-server-spotinst

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
1
Forks
0
Acompanhando
0
Saúde
42%
Último commit
há 6 meses
Versão
v0.7.0
Issues abertas
0
Contribuidores
2

Casos de uso

  • Monitorar e gerenciar clusters Kubernetes Ocean em múltiplas contas AWS/Azure
  • Analisar custos de infraestrutura e tendências week-over-week
  • Executar operações de scaling, rolling updates e provisionamento de nós
  • Consultar recomendações de right-sizing de recursos
  • Exportar configurações de cluster em YAML para GitOps/backup

Veredito

MCP server robusto e bem-estruturado para equipes DevOps/SRE que usam Spotinst em produção. Oferece 34 ferramentas cobrindo operações leitura/escrita, análise de custos e multi-cloud, com confirmação obrigatória para operações destrutivas — ideal para automação e consulta em ambientes AWS/Azure complexos.

README

A tradução para o português deste README ainda está na fila.

mcp-server-spotinst

PyPI version CI License: MIT Python 3.10+

MCP server for the Spot.io (Spotinst) API. Supports both AWS and Azure Ocean clusters with multi-account access.

Tools (34)

Cross-Account

ToolDescription
list_accountsList all Spotinst accounts accessible with the current token
list_all_clustersList ALL clusters across ALL accounts and clouds (AWS + Azure)

AWS Ocean

ToolDescription
list_clustersList AWS Ocean K8s clusters
get_clusterGet AWS Ocean cluster details
list_vngsList AWS Virtual Node Groups (launch specs)
get_vngGet AWS VNG details
list_elastigroupsList all Elastigroups
get_elastigroupGet Elastigroup details
get_allowed_instance_typesGet allowed EC2 instance types
get_right_sizingGet right-sizing resource suggestions (AWS only)

Azure Ocean

ToolDescription
list_clusters_azureList Azure Ocean clusters
get_cluster_azureGet Azure Ocean cluster details
list_vngs_azureList Azure Virtual Node Groups
get_vng_azureGet Azure VNG details

Both Clouds (pass cloud="azure" for Azure)

ToolDescription
get_cluster_nodesList nodes in an Ocean cluster
get_cluster_costsGet aggregated cost breakdown by namespace or resource
list_rollsList deployment rolls
get_rollGet roll details
get_cluster_logGet scaling and activity log events

Stateful Nodes (AWS Managed Instances)

ToolDescription
list_stateful_nodesList all Stateful Nodes (Managed Instances)
get_stateful_nodeGet Stateful Node details

Scheduling & Health

ToolDescription
get_cluster_schedulingGet scheduling and auto-scaler configuration
get_cluster_healthComposite health check: nodes, recent errors, active rolls

Cost Analysis

ToolDescription
get_cost_trendingWeek-over-week (or custom period) cost comparison
get_savings_summary30-day cost and savings summary

Tag Filtering

ToolDescription
filter_clusters_by_tagFilter clusters by tag key/value
filter_vngs_by_tagFilter VNGs by tag key/value

Export (YAML)

ToolDescription
export_cluster_yamlExport cluster config as YAML for GitOps/backup
export_vng_yamlExport VNG config as YAML for GitOps/backup

Write Operations (require confirm=true)

ToolDescription
remove_instancesRecommended — Remove instances using a named strategy (see below)
initiate_rollRolling restart of nodes in an Ocean cluster
detach_instancesDetach and optionally terminate instances from an AWS Ocean cluster
update_vngUpdate an AWS VNG configuration
update_vng_azureUpdate an Azure VNG configuration

All tools accept an optional account_id parameter to query any account.

Instance Removal Strategies

The remove_instances tool provides a safe, intent-based interface for removing instances. Instead of remembering which API flags to set, you pick a strategy:

StrategyWhat happensUse case
drain_and_replaceGracefully drain pods (respects PDBs), terminate, Ocean replaces. Default and safest.Replacing a problematic node in production
replaceImmediately terminate, Ocean auto-replaces. No graceful drain. AWS only.Fast replacement when drain isn't needed
remove_permanentlyTerminate + reduce cluster capacity. No replacement. AWS only.Downsizing the cluster

When confirm=false (default), the tool shows a detailed execution plan so you can review before proceeding:

remove_instances("o-abc123", "i-abc123", strategy="drain_and_replace")

→ SAFETY: Action NOT executed. Set confirm=true to proceed.
  DRAIN AND REPLACE 1 instance(s) in cluster o-abc123:
    Instances: ['i-abc123']
    Method: Rolling restart (20% per batch)
    - Pods will be gracefully drained (PDBs respected)
    - Instances will be terminated after drain
    - Ocean will automatically launch replacements

Setup

Environment Variables

export SPOTINST_TOKEN="your-spotinst-api-token"
export SPOTINST_ACCOUNT_ID="act-xxxxxxxx"

Install with pip

pip install mcp-server-spotinst

Install with uvx (no install needed)

uvx mcp-server-spotinst

Claude Code Config

Add to ~/.mcp.json:

{
  "mcpServers": {
    "spotinst": {
      "command": "uvx",
      "args": ["mcp-server-spotinst"],
      "env": {
        "SPOTINST_TOKEN": "your-token",
        "SPOTINST_ACCOUNT_ID": "act-xxxxxxxx"
      }
    }
  }
}

Docker

docker build -t mcp-server-spotinst .
docker run -e SPOTINST_TOKEN=your-token -e SPOTINST_ACCOUNT_ID=act-xxxxxxxx mcp-server-spotinst

Run Standalone

mcp-server-spotinst

Multi-Account Usage

Your personal API token can access multiple Spotinst accounts. Use list_accounts to see all available accounts, then pass account_id to any tool:

list_clusters(account_id="act-be5e7ffe")
get_cluster_nodes("o-390ef886", account_id="act-9785011e", cloud="azure")

Or use list_all_clusters for a single-call inventory across everything.

API Reference

Uses the Spot.io REST API at https://api.spotinst.io.

Ver o original no GitHub

Bom para

Relacionados