Cloud Manage MCP Server
Provides comprehensive management of DigitalOcean droplets, including power management (start, stop, reboot), status monitoring, operation history tracking, batch management, resource usage metrics, and deletion protection mechanisms.
Utilizes environment variables for configuration management, allowing secure storage of API tokens and feature flags through a .env file.
Supports management of Ubuntu-based droplets, with specific handling for Ubuntu operating system distributions on DigitalOcean.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Cloud Manage MCP Server list all my droplets"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Cloud Manage MCP Server
这是一个基于 MCP (Model Context Protocol) 的云管理服务器,支持通过公网 IP 地址获取云服务器信息。
功能特性
支持通过公网 IP 地址识别云服务提供商
使用 pydo SDK 获取 DigitalOcean droplet 详细信息
Droplet 电源管理: 开机、关机、重启、优雅关机
状态监控: 获取 droplet 当前状态和资源使用情况
操作历史: 查看 droplet 操作历史记录
批量管理: 列出所有 droplets,按名称查找
监控数据: 获取 CPU、内存、磁盘、网络使用率(需启用监控)
删除保护: 多重安全机制防止意外删除重要服务器
支持 AWS(待实现)
Related MCP server: Hetzner Cloud MCP Server
安全特性
🛡️ 删除保护机制
为了防止意外删除重要的 droplet,本系统实现了严格的五层安全保护:
保护层级
全局开关保护: 默认禁用所有删除操作
确认码保护: 需要特定确认码才能尝试删除
状态检查保护: 禁止删除运行中的 droplet
标签保护: 自动识别并保护重要 droplet
最终安全检查: 即使通过前面检查,实际删除仍被禁用
保护标签
系统会自动保护带有以下标签的 droplet:
production/prodimportantcriticalbackup
配置方法
# 默认情况下,删除功能完全禁用
# 如需启用(强烈不推荐),设置环境变量:
export ALLOW_DROPLET_DELETION=true⚠️ 重要提醒: 即使启用删除功能,所有实际删除操作仍会被最终安全检查阻止。建议始终通过 DigitalOcean 控制面板手动删除 droplet。
可用的 MCP 工具函数
基础查询功能
get_dg_info(ipv4_address: str)
根据公网 IP 地址获取对应的 DigitalOcean droplet 信息。
参数:
ipv4_address(str): 要查询的公网 IP 地址
list_droplets()
列出账户下所有的 DigitalOcean droplets。
find_droplet_by_name(name: str)
根据名称查找 DigitalOcean droplet(支持模糊匹配)。
参数:
name(str): droplet 名称或部分名称
get_droplet_status(droplet_id: int)
获取指定 droplet 的当前状态和详细信息。
参数:
droplet_id(int): droplet ID
电源管理功能
power_on_droplet(droplet_id: int)
开启指定的 droplet。
power_off_droplet(droplet_id: int)
强制关闭指定的 droplet(类似拔电源)。
shutdown_droplet(droplet_id: int)
优雅关闭指定的 droplet(类似系统关机命令)。
reboot_droplet(droplet_id: int)
重启指定的 droplet。
参数:
droplet_id(int): 要操作的 droplet ID
监控和历史功能
get_droplet_monitoring(droplet_id: int)
获取 droplet 的监控数据,包括 CPU、内存、磁盘、网络使用率。
注意: 需要在 droplet 上启用监控功能。
get_droplet_actions(droplet_id: int)
获取指定 droplet 的操作历史记录。
get_action_status(action_id: int)
查询特定操作的状态(在执行电源操作后可用于跟踪进度)。
参数:
droplet_id(int): droplet IDaction_id(int): 操作 ID
删除保护功能
delete_droplet_with_protection(droplet_id: int, confirmation_code: str = "")
删除 DigitalOcean droplet(带严格安全保护)。
注意: 此功能具有多重安全限制,实际上被设计为阻止删除操作。
参数:
droplet_id(int): 要删除的 droplet IDconfirmation_code(str): 确认码(必须为 "CONFIRM_DELETE_DROPLET")
get_droplet_deletion_policy()
获取当前的删除策略和安全配置信息。
check_droplet_deletion_safety(droplet_id: int)
检查指定 droplet 的删除安全性(不实际删除)。
参数:
droplet_id(int): 要检查的 droplet ID
使用示例
基本操作示例
# 列出所有 droplets
result = list_droplets()
print(f"总共有 {result['total_droplets']} 个 droplets")
# 查找特定名称的 droplet
result = find_droplet_by_name("web-server")
if result['found']:
droplet = result['droplets'][0]
droplet_id = droplet['id']
# 获取详细状态
status = get_droplet_status(droplet_id)
print(f"Droplet 状态: {status['status']}")
# 重启 droplet
if status['status'] == 'active':
action_result = reboot_droplet(droplet_id)
action_id = action_result['action']['id']
# 检查操作状态
action_status = get_action_status(action_id)
print(f"重启操作状态: {action_status['action']['status']}")删除安全性检查示例
# 检查删除策略
policy = get_droplet_deletion_policy()
print(f"删除功能状态: {policy['deletion_policy']['current_status']}")
# 检查特定 droplet 的删除安全性
safety_check = check_droplet_deletion_safety(droplet_id)
print(f"安全等级: {safety_check['safety_level']}")
# 查看安全检查详情
for check in safety_check['safety_checks']:
print(f"{check['check']}: {check['status']} - {check['message']}")
# 尝试删除(会被安全机制阻止)
deletion_result = delete_droplet_with_protection(droplet_id, "CONFIRM_DELETE_DROPLET")
if deletion_result.get('error'):
print(f"删除被阻止: {deletion_result['error']}")监控数据获取示例
# 获取监控数据
monitoring_result = get_droplet_monitoring(droplet_id)
if monitoring_result['monitoring_enabled']:
metrics = monitoring_result['metrics']
for metric_type, data in metrics.items():
if data['available']:
print(f"{metric_type} 监控数据可用")
else:
print(f"{metric_type} 暂无数据")
else:
print("请先在 DigitalOcean 控制面板中启用监控功能")返回值说明
成功响应示例
删除安全性检查
{
"cloud_provider": "digitalocean",
"droplet_id": 123456789,
"droplet_name": "web-server-01",
"overall_safety": "BLOCKED",
"safety_level": "删除被阻止",
"safety_checks": [
{
"check": "全局删除策略",
"status": "BLOCKED",
"message": "删除功能已被全局禁用"
},
{
"check": "droplet状态",
"status": "WARNING",
"message": "droplet正在运行 (active),建议先关机"
},
{
"check": "保护标签",
"status": "BLOCKED",
"message": "发现保护标签: production"
}
],
"warnings": [
"此droplet带有保护标签,表明它可能是重要服务器"
],
"summary": {
"total_checks": 5,
"blocked": 2,
"warnings": 1,
"passed": 2
}
}删除策略信息
{
"cloud_provider": "digitalocean",
"deletion_policy": {
"enabled": false,
"protection_level": "MAXIMUM",
"current_status": "所有删除操作被禁用",
"safety_checks": [
"环境变量 ALLOW_DROPLET_DELETION 必须设置为 true",
"必须提供正确的确认码 'CONFIRM_DELETE_DROPLET'",
"droplet 必须处于关机状态",
"droplet 不能带有保护标签",
"多重确认机制"
],
"protected_tags": ["production", "prod", "important", "critical", "backup"]
},
"security_info": {
"philosophy": "安全第一,防止意外删除重要服务器",
"recommendation": "强烈建议通过DigitalOcean控制面板手动删除droplet"
}
}droplet 操作成功
{
"cloud_provider": "digitalocean",
"droplet_id": 123456789,
"action": {
"id": 987654321,
"status": "in-progress",
"type": "reboot",
"started_at": "2024-01-01T12:00:00Z",
"completed_at": null,
"resource_id": 123456789,
"resource_type": "droplet",
"region": "New York 3"
},
"message": "已成功提交 reboot 操作,操作ID: 987654321"
}droplet 状态查询
{
"cloud_provider": "digitalocean",
"droplet_id": 123456789,
"status": "active",
"name": "web-server-01",
"locked": false,
"size_slug": "s-1vcpu-1gb",
"memory": 1024,
"vcpus": 1,
"disk": 25,
"region": {
"name": "New York 3",
"slug": "nyc3"
},
"image": {
"name": "Ubuntu 20.04 x64",
"distribution": "Ubuntu"
},
"created_at": "2024-01-01T10:00:00Z",
"features": ["monitoring", "ipv6"],
"tags": ["web", "production"]
}监控数据响应
{
"cloud_provider": "digitalocean",
"droplet_id": 123456789,
"monitoring_enabled": true,
"metrics": {
"cpu": {
"available": true,
"data": [...]
},
"memory": {
"available": true,
"data": [...]
},
"disk": {
"available": false,
"data": []
},
"network": {
"available": true,
"data": [...]
}
},
"note": "监控数据可能需要几分钟才能在新启用监控的droplet上显示。"
}错误响应示例
删除操作被阻止
{
"cloud_provider": "digitalocean",
"error": "删除操作已被禁用。出于安全考虑,所有droplet删除操作被限制。",
"security_info": {
"protection_level": "MAXIMUM",
"reason": "防止意外删除重要服务器",
"how_to_enable": "设置环境变量 ALLOW_DROPLET_DELETION=true (不推荐)"
}
}一般错误
{
"cloud_provider": "digitalocean",
"error": "未找到ID为 123456789 的droplet"
}测试功能
项目包含一个完整的测试脚本 test_dg_info.py,可以测试所有功能:
python test_dg_info.py测试脚本包含以下测试模块:
基本功能测试: 列出 droplets、API 连接测试
Droplet 操作测试: 状态查询、电源管理、操作历史
监控功能测试: 监控数据获取和可用性检查
删除保护测试: 安全策略检查、删除安全性评估
IP 查找测试: 根据 IP 地址查找对应的 droplet
注意事项
安全相关
删除保护:
默认禁用所有删除操作以防止意外删除
多重安全检查机制保护重要服务器
即使配置允许删除,实际删除仍被最终安全检查阻止
强烈建议通过 DigitalOcean 控制面板手动删除
API 权限: 确保您的 DigitalOcean API token 具有适当的权限:
读取 droplets 的权限
执行 droplet 操作的权限(开关机、重启等)
访问监控数据的权限
监控功能限制:
需要在 droplet 上启用监控功能(在创建时或后续在控制面板中启用)
监控数据可能需要几分钟才能显示
免费 monitoring 功能每 5 分钟收集一次数据点
API 调用限制:
DigitalOcean API 有频率限制(每小时 5,000 次请求)
电源操作是异步的,需要查询操作状态来确认完成
安全考虑:
请妥善保管您的 API token
建议使用环境变量存储敏感信息
在生产环境中限制 API token 的权限范围
避免在自动化脚本中启用删除功能
操作状态说明
droplet 可能的状态包括:
new: 新创建,正在初始化active: 运行中off: 已关机archive: 已归档(长期关闭)
操作状态包括:
in-progress: 执行中completed: 已完成errored: 执行失败
删除安全等级:
BLOCKED: 删除被阻止(有严重安全问题)WARNING: 需要谨慎考虑(有潜在风险)CAUTION: 可以删除但需确认(仍有限制)
最佳实践
安全操作:
避免启用删除功能,始终通过控制面板手动删除
定期使用
check_droplet_deletion_safety()评估服务器安全性为重要服务器添加保护标签
批量操作: 使用
list_droplets()获取所有 droplets,然后根据需要过滤和操作状态检查: 在执行电源操作前,先检查 droplet 当前状态
操作跟踪: 保存操作 ID,用于后续状态查询
错误处理: 始终检查返回结果中的
error字段监控启用: 在创建 droplet 时启用监控功能,以便后续获取使用数据
依赖安装
pip install -r requirements.txt环境变量配置
创建 .env 文件并设置以下环境变量:
# DigitalOcean API Token
# 在 https://cloud.digitalocean.com/account/api/tokens 获取
DIGITALOCEAN_TOKEN=your_digitalocean_api_token_here
# IPInfo API Token (可选,用于IP地址查询)
# 在 https://ipinfo.io/account/token 获取
IPINFO_API_TOKEN=your_ipinfo_api_token_here
# 删除功能控制 (强烈不推荐启用)
# ALLOW_DROPLET_DELETION=false # 默认值,建议保持禁用使用方法
启动服务器
python main.pyAPI 说明
pydo SDK 集成
本项目使用官方的 pydo SDK 来与 DigitalOcean API 交互:
安装 pydo:
pip install pydo配置 API Token: 设置
DIGITALOCEAN_TOKEN环境变量调用 API: 使用
client.droplets.list()获取所有 droplets
工作原理
初始化 pydo 客户端,使用配置的 API token
调用
client.droplets.list()获取账户下所有 droplets遍历每个 droplet 的网络配置
查找匹配指定公网 IP 地址的 droplet
返回找到的 droplet 的详细信息
错误处理
自动检测 pydo SDK 是否已安装
验证 DIGITALOCEAN_TOKEN 环境变量是否配置
捕获 API 调用异常并返回友好的错误信息
提供详细的调试信息
注意事项
确保您的 DigitalOcean API token 具有读取 droplets 的权限
API 有调用频率限制,请合理使用
该功能会列出账户下所有 droplets,请确保 token 的安全性
扩展功能
支持按标签过滤 droplets
支持分页查询大量 droplets
缓存查询结果以提高性能
支持其他云服务提供商(AWS、GCP 等)
Available Tools
30 toolscheck_provider_availabilityC
检查特定云服务提供商的可用性
Args:
provider_name (str): 提供商名称 ('aws', 'digitalocean', 'vultr', 'alibaba')
Returns:
Dict: 提供商可用性信息
Inputs: provider_name (required). Returns: result
get_alibaba_instance_infoC
获取阿里云ECS实例信息
Args:
ip_address_or_id (str): 公网IP地址或实例ID
Returns:
Dict: 阿里云实例信息
Inputs: ip_address_or_id (required). Returns: result
get_alibaba_instance_monitoringC
获取阿里云ECS实例监控信息
Inputs: instance_id (required). Returns: result
get_aws_instance_infoB
获取AWS EC2实例信息(只读)
Args:
ip_address_or_id (str): 公网IP地址或实例ID
Returns:
Dict: AWS实例信息
Inputs: ip_address_or_id (required). Returns: result
get_aws_instance_monitoringC
获取AWS EC2实例的监控数据
Args:
instance_id (str): EC2实例ID
hours (int): 获取过去多少小时的数据
Returns:
Dict: 监控数据
Inputs: instance_id (required), hours. Returns: result
get_aws_instance_storage_infoB
获取AWS EC2实例的存储详细信息
Args:
instance_id (str): EC2实例ID
Returns:
Dict: 存储信息,包括磁盘类型、IOPS、吞吐量等
Inputs: instance_id (required). Returns: result
get_digitalocean_droplet_actionsC
获取DigitalOcean Droplet操作历史
Inputs: droplet_id (required). Returns: result
get_digitalocean_droplet_infoB
获取DigitalOcean Droplet信息
Args:
ip_address_or_id (str): 公网IP地址或Droplet ID
Returns:
Dict: Droplet信息
Inputs: ip_address_or_id (required). Returns: result
get_digitalocean_droplet_monitoringC
获取DigitalOcean Droplet监控信息
Inputs: droplet_id (required). Returns: result
get_instance_by_providerA
通过明确指定的云服务提供商查询实例信息
Args:
provider (str): 云服务提供商 ('aws', 'digitalocean', 'vultr', 'alibaba')
identifier (str): 实例标识符(IP地址或实例ID)
Returns:
Dict: 实例信息
Inputs: provider (required), identifier (required). Returns: result
get_instance_infoA
根据IP地址自动检测云服务提供商并获取实例信息
Args:
ip_address (str): 公网IP地址
provider (str, optional): 明确指定的云服务提供商 ('aws', 'digitalocean', 'vultr', 'alibaba')
Returns:
Dict: 实例信息,包含提供商信息和实例详情
Inputs: ip_address (required), provider. Returns: result
get_supported_providersB
获取支持的云服务提供商列表
Returns:
Dict: 支持的云服务提供商信息
Inputs: no named parameters. Returns: result
get_system_statusB
获取整个系统的状态概览
Returns:
Dict: 系统状态信息
Inputs: no named parameters. Returns: result
get_vultr_instance_bandwidthC
获取Vultr实例带宽使用情况
Inputs: instance_id (required). Returns: result
get_vultr_instance_infoC
获取Vultr实例信息
Args:
ip_address_or_id (str): 公网IP地址或实例ID
Returns:
Dict: Vultr实例信息
Inputs: ip_address_or_id (required). Returns: result
list_alibaba_instancesB
列出所有阿里云ECS实例
Inputs: no named parameters. Returns: result
list_aws_instancesB
列出所有AWS EC2实例
Returns:
Dict: AWS实例列表
Inputs: no named parameters. Returns: result
list_digitalocean_dropletsB
列出所有DigitalOcean Droplets
Inputs: no named parameters. Returns: result
list_vultr_instancesC
列出所有Vultr实例
Inputs: no named parameters. Returns: result
manage_instance_powerB
通用的实例电源管理函数(支持所有云平台)
Args:
provider (str): 云服务提供商 ('digitalocean', 'vultr', 'alibaba')
instance_id (str): 实例ID
action (str): 操作类型 ('power_on', 'power_off', 'reboot', 'shutdown')
ip_confirmation (str): 确认IP地址
name_confirmation (str): 确认实例名称
operation_confirmation (str): 确认操作类型
Returns:
Dict: 操作结果
Inputs: provider (required), instance_id (required), action (required), ip_confirmation, +2 more. Returns: result
power_off_alibaba_instanceC
强制停止阿里云ECS实例(需要三次确认)
Inputs: instance_id (required), ip_confirmation, name_confirmation, operation_confirmation. Returns: result
power_off_digitalocean_dropletC
强制关闭DigitalOcean Droplet(需要三次确认)
Inputs: droplet_id (required), ip_confirmation, name_confirmation, operation_confirmation. Returns: result
power_off_vultr_instanceC
强制关闭Vultr实例(需要三次确认)
Inputs: instance_id (required), ip_confirmation, name_confirmation, operation_confirmation. Returns: result
power_on_alibaba_instanceC
启动阿里云ECS实例(需要三次确认)
Inputs: instance_id (required), ip_confirmation, name_confirmation, operation_confirmation. Returns: result
power_on_digitalocean_dropletA
开启DigitalOcean Droplet(需要三次确认)
Args:
droplet_id (int): Droplet ID
ip_confirmation (str): 确认IP地址
name_confirmation (str): 确认Droplet名称
operation_confirmation (str): 确认操作类型(输入"开机"或"power_on")
Returns:
Dict: 操作结果或确认要求
Inputs: droplet_id (required), ip_confirmation, name_confirmation, operation_confirmation. Returns: result
power_on_vultr_instanceB
开启Vultr实例(需要三次确认)
Inputs: instance_id (required), ip_confirmation, name_confirmation, operation_confirmation. Returns: result
reboot_alibaba_instanceC
重启阿里云ECS实例(需要三次确认)
Inputs: instance_id (required), ip_confirmation, name_confirmation, operation_confirmation. Returns: result
reboot_digitalocean_dropletC
重启DigitalOcean Droplet(需要三次确认)
Inputs: droplet_id (required), ip_confirmation, name_confirmation, operation_confirmation. Returns: result
reboot_vultr_instanceC
重启Vultr实例(需要三次确认)
Inputs: instance_id (required), ip_confirmation, name_confirmation, operation_confirmation. Returns: result
shutdown_digitalocean_dropletC
优雅关闭DigitalOcean Droplet(需要三次确认)
Inputs: droplet_id (required), ip_confirmation, name_confirmation, operation_confirmation. Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 30 tools
There is significant overlap and unclear boundaries between tools. For example, get_instance_by_provider, get_instance_info, and individual provider-specific get_*_instance_info tools all retrieve instance information, making it ambiguous which to use. Similarly, power management tools are duplicated across providers (e.g., power_on_alibaba_instance, power_on_digitalocean_droplet) while a generic manage_instance_power also exists, causing confusion. The descriptions do not clearly differentiate when to use generic vs. provider-specific tools.
The naming follows a consistent verb_noun pattern with minor deviations. Most tools use snake_case and start with verbs like get, list, manage, power_off, etc., followed by provider and resource names (e.g., get_aws_instance_info, list_digitalocean_droplets). However, there are inconsistencies like check_provider_availability (which doesn't fit the provider-specific pattern) and some tools omit provider names in verbs (e.g., get_system_status, get_supported_providers), but overall the naming is mostly predictable.
With 30 tools, the count is too high for the apparent scope of cloud instance management across four providers. The server includes many redundant tools (e.g., separate power management tools for each provider instead of consolidating) and could be streamlined. This large number increases complexity and potential for confusion, making it feel heavy and over-engineered for the domain.
The tool surface covers core operations like listing instances, getting info, monitoring, and power management across providers, but there are notable gaps. For example, there are no tools for creating, updating, or deleting instances, which are essential CRUD operations in cloud management. Additionally, while monitoring and bandwidth tools exist for some providers, coverage is inconsistent (e.g., missing for Alibaba in descriptions). The surface allows basic read and power control but lacks full lifecycle management.
Maintenance
Related MCP Connectors
An MCP server that provides read access to your cloud storage providers, bank accounts and more.
The Telnyx MCP server is an official implementation of the Model Context Protocol that enables AI clients (like Claude Desktop, Cursor, and OpenAI Agents) to interact with Telnyx's telephony, messaging, and AI assistant APIs. It provides comprehensive capabilities including making and managing phone calls, sending SMS/MMS messages, purchasing and configuring phone numbers, creating AI assistants with custom instructions, managing cloud storage buckets, scraping and embedding website content, and handling integration secrets. The server exists as both a local implementation and a remotely hosted version, allowing developers to integrate real-world communication infrastructure directly into AI applications.
An MCP server that provides an API to LLMs to manage their JumpCloud resources.
A Model Context Protocol server for Wix AI tools
Related MCP Servers
- AlicenseAqualityFmaintenanceA Model Context Protocol server that retrieves detailed geographic information about IP addresses using the ipinfo.io API, helping determine user location and network details.560 PyPI45MIT
- AlicenseAqualityDmaintenanceA Model Context Protocol server that allows language models to manage Hetzner Cloud resources through structured functions, including servers, volumes, firewalls, and SSH keys.30153MIT
- AlicenseNot gradedqualityDmaintenanceA server that implements the Model Context Protocol, providing a standardized way to connect AI models to different data sources and tools.12MIT
- AlicenseNot gradedqualityDmaintenanceA Model Context Protocol (MCP) server that provides programmatic access to DigitalOcean's API. This server exposes tools for managing droplets, Kubernetes clusters, and container registries through the MCP interface.4MIT