Alibaba Cloud MCP Server
by
Provides seamless integration of Alibaba Cloud services into AI applications via Model Context Protocol, exposing tens of thousands of OpenAPI endpoints for rapid workflow composition and automation.
Alibaba Cloud MCP Server Overview
What is Alibaba Cloud MCP Server about?
Alibaba Cloud MCP Server enables AI applications to call a comprehensive catalog of Alibaba Cloud OpenAPI services without writing low‑level request code. It abstracts the underlying REST APIs, supports remote (cloud‑hosted) and local (stdio‑based) deployment modes, and delivers AI‑friendly API definitions, multi‑account handling, and custom OAuth control.
How to use Alibaba Cloud MCP Server?
Remote Mode
- Open the appropriate endpoint:
- China mainland:
https://api.aliyun.com/mcp - International:
https://api.alibabacloud.com/mcp
- China mainland:
- Register an OAuth 2.0 application in the Alibaba Cloud console, obtain Access Token and configure token‑refresh logic.
- Browse the system MCP service list (Infrastructure, Monitoring, Governance, etc.) and select the desired OpenAPI.
- Optionally customize API descriptions, simplify optional parameters, and adjust token lifetimes.
Local Mode
- Clone the target service repository, e.g.:
git clone https://github.com/aliyun/alibabacloud-rds-openapi-mcp-server cd alibabacloud-rds-openapi-mcp-server - Install dependencies:
npm install # or yarn install - Export credentials as environment variables:
export ALIBABA_CLOUD_ACCESS_KEY_ID=your_access_key_id export ALIBABA_CLOUD_ACCESS_KEY_SECRET=your_access_key_secret - Start the MCP server (stdio mode):
The process communicates with AI clients over standard input/output streams.npm start
Key Features
- OpenAPI Customization: Refine API descriptions, hide non‑essential parameters, and improve AI parsing accuracy.
- Terraform as Tools: Directly expose Terraform HCL definitions, auto‑convert variables to tool parameters, and ensure deterministic infrastructure orchestration.
- Multi‑Account Support: Role‑assumption, account switching, and centralized management for cross‑account AI workflows.
- Custom OAuth: Callback whitelist, configurable token lifetimes, and up‑to‑1‑year login‑free tokens.
- Remote & Local Deployments: Zero‑maintenance cloud‑hosted service or self‑hosted stdio process for strict data‑security requirements.
- System MCP Service Catalog: Includes Infrastructure Management (Terraform Provider, Quota Center, Resource Search), Monitoring & Auditing (AI‑driven Audit, Permission Diagnosis), and Governance (Compliance Packages, Governance Reports).
Use Cases
- AI‑Driven Cloud Automation: Let large language models provision, configure, and monitor resources without writing SDK code.
- Compliance & Governance Bots: Automatically query governance reports or audit logs through conversational interfaces.
- Multi‑Tenant SaaS Platforms: Manage resources across dozens of customer accounts via role‑assumption, all driven by AI agents.
- Infrastructure as Code Integration: Convert Terraform scripts into AI‑callable tools for dynamic, on‑the‑fly provisioning.
- Secure Edge Deployments: Run the MCP server locally in isolated environments (e.g., on‑prem data centers) while still accessing Alibaba Cloud services.
FAQ
Q: Do I need to run a full server for every Alibaba Cloud service? A: No. The remote mode provides a single, cloud‑hosted MCP endpoint covering all public OpenAPI services. Local mode is optional for specific services that require custom credentials or offline operation.
Q: How many OpenAPI endpoints are supported? A: Tens of thousands of Alibaba Cloud OpenAPI definitions are available out‑of‑the‑box.
Q: Can I use my own OAuth provider? A: Yes. The platform allows custom OAuth configurations, including callback white‑listing and token lifetime settings.
Q: What languages can interact with the MCP server? A: Any language that can issue HTTP/HTTPS requests or communicate via stdio (e.g., Python, JavaScript, Java, Go) can integrate.
Q: Is there a CLI for quick testing?
A: The remote CLI mode is bundled with the core tools; you can invoke it via npx as shown in the serverConfig section.
Alibaba Cloud MCP Server's README
阿里云 MCP Server 使用指南
README of English
📚 目录
概述
阿里云 MCP Server 是一个强大的云服务集成平台，通过 Model Context Protocol (MCP) 为 AI 应用提供阿里云服务的无缝集成能力。该平台支持数万个阿里云 OpenAPI，让开发者能够轻松地将阿里云的各种服务能力集成到 AI 工作流中。
部署模式
远程模式
通过阿里云官方托管的 OpenAPI MCP Server，无需本地部署即可使用全量阿里云服务。适合快速集成、低维护成本的场景。
阿里云现已提供 API MCP Server Core。该模式基于远程 CLI 模式运行，通过个位数的核心 tools 即可编排并触达阿里云全量能力（覆盖数万 OpenAPI）。
本地模式
基于 stdio 进程模式，在本地运行 MCP Server。适合对数据安全性要求高、需要自定义配置的场景。
远程模式详解
技术规格
- 核心模式：API MCP Server Core（远程 CLI 模式）
- 支持协议：SSE (Server-Sent Events)、Streamable HTTP
- 认证方式：OAuth 2.0
- API 数量：支持阿里云数万个 OpenAPI
- 部署方式：云端托管，零运维
访问地址
| 区域 | 访问地址 | 适用场景 |
|---|---|---|
| 中国站 | https://api.aliyun.com/mcp | 中国大陆用户 |
| 国际站 | https://api.alibabacloud.com/mcp | 海外及国际用户 |
系统 MCP 服务列表
阿里云官方提供了一系列经过精心调优的系统 MCP 服务，针对特定场景进行了优化：
基础设施管理
| 服务名称 | 功能描述 |
|---|---|
| Terraform Provider | 提供阿里云 Terraform Provider 元数据，支持在线验证和执行 Terraform 命令的 Runtime 能力 |
| 配额中心 | 根据云产品名称、配额描述、地域信息等，查询配额中心支持的产品通用配额信息 |
| 资源搜索 | 支持当前账号下有权限资源的搜索和统计功能 |
监控与审计
| 服务名称 | 功能描述 |
|---|---|
| 操作审计 AI | 使用 AI 根据场景灵活调用操作审计的 LookupEvents 接口 |
| 权限诊断 | API 请求因无权限被拒绝时，通过 EncodedDiagnosticMessage 进行权限诊断 |
合规与治理
| 服务名称 | 功能描述 |
|---|---|
| 治理报告 | 基于 GovernanceReport 的 MCP Server |
| 配置审计合规包 | 查询合规包模板、启用指定合规包、查询风险项概况及风险资源清单 |
核心能力
OpenAPI 定制化调优
- 修改 API 描述，使其更适合 AI 理解
- 精简非必填参数，降低调用复杂度
- 优化参数说明，提高 AI 调用准确率
Terraform As Tools
- HCL 代码集成：将 Terraform HCL 代码作为完整工具引入
- 变量自动解析：Terraform 变量自动转换为工具参数
- 确定性编排：实现基础设施的确定性部署和管理
多账号支持
- 角色扮演：自动使用角色扮演能力操作特定账号
- 账号切换：灵活指定操作账号和扮演角色
- 集中管理：轻松实现多账号 AI 集成管理
自定义 OAuth
- Callback 白名单：精确控制回调地址白名单
- Token 生命周期：灵活设置 access token 和 refresh token 过期时间
- 长期免登录：最长可实现 1 年免登录
本地模式详解
运行机制
本地模式基于 stdio 进程通信，MCP Server 作为独立进程在本地运行，通过标准输入输出与 AI 应用进行通信。
服务列表
数据库服务
| 服务 | 仓库地址 | 描述 |
|---|---|---|
| DMS | alibabacloud-dms-mcp-server | 数据管理服务，提供数据库管理能力 |
| RDS | alibabacloud-rds-openapi-mcp-server | 关系型数据库服务 OpenAPI 集成 |
| ADBPG | alibabacloud-adbpg-mcp-server | 分析型数据库 PostgreSQL 版 |
数据分析服务
| 服务 | 仓库地址 | 描述 |
|---|---|---|
| DataWorks | alibabacloud-dataworks-mcp-server | 数据工场，提供大数据开发治理能力 |
DevOps 服务
| 服务 | 仓库地址 | 描述 |
|---|---|---|
| 云效 | alibabacloud-devops-mcp-server | 企业级 DevOps 平台集成 |
| 运维开发 | alibaba-cloud-ops-mcp-server | 运维开发工具集成 |
其他服务
| 服务 | 仓库地址 | 描述 |
|---|---|---|
| ESA | mcp-server-esa | 边缘安全加速服务 |
| 可观测 | alibabacloud-observability-mcp-server | 可观测性服务集成 |
快速开始
远程模式接入
-
访问控制台
- 中国站用户访问：https://api.aliyun.com/mcp
- 国际站用户访问：https://api.alibabacloud.com/mcp
-
OAuth 认证
- 配置 OAuth 应用
- 获取 Access Token
- 配置 Token 刷新策略
-
选择服务
- 浏览系统 MCP 服务
- 选择需要的 OpenAPI
- 自定义 API 参数
通过本地静态凭证接入
现在阿里云 API MCP Server 可以通过本地静态凭证直接登录。你可以在本地配置阿里云 AccessKey 或已有凭证文件，然后通过 Alibaba Cloud MCP Proxy 自动换取访问 OpenAPI MCP Server 所需的令牌，无需在 MCP 客户端中手动维护 OAuth Token。
代理工具的安装、MCP 客户端配置、安全策略和预检查说明请参考：Alibaba Cloud MCP Proxy 使用说明。
本地模式部署
-
克隆仓库
git clone https://github.com/aliyun/[具体服务仓库名称] cd [仓库目录] -
安装依赖
npm install # 或 yarn install -
配置认证
export ALIBABA_CLOUD_ACCESS_KEY_ID=your_access_key export ALIBABA_CLOUD_ACCESS_KEY_SECRET=your_secret_key -
启动服务
npm start # 或按照具体仓库的启动说明
参考文档
- 📖 官方文档：OpenAPI MCP Server 使用指南
- 🔧 技术支持：通过阿里云工单系统或官方论坛获取技术支持
- 💬 社区交流：加入阿里云开发者社区参与讨论，钉钉群：136325002292
- 通过AgentScope使用示例https://github.com/agentscope-ai/agentscope/tree/main/examples/alibabacloud_api_mcp
本文档持续更新中，欢迎提交 Issue 或 PR 贡献内容
Alibaba Cloud MCP Server Reviews
Login Required
Please log in to share your review and rating for this MCP.
Similar MCP Servers like Alibaba Cloud MCP Server
Explore related MCPs that share similar capabilities and solve comparable challenges
Aws Mcp Servers
Officialby awslabs
Provides specialized servers that expose AWS capabilities through the Model Context Protocol, enabling AI assistants to retrieve up-to-date documentation, execute API calls, and automate infrastructure workflows directly within development environments.
Cloudflare MCP Server
by cloudflare
Provides a collection of Model Context Protocol servers that enable MCP‑compatible clients to interact with Cloudflare services such as Workers, Observability, Radar, and more, allowing natural‑language driven management of configurations, data, and operations.
MCP Server Kubernetes
by Flux159
Connects to a Kubernetes cluster and offers a unified MCP interface for kubectl, Helm, port‑forwarding, diagnostics, and non‑destructive read‑only mode.
Run Model Context Protocol Servers With AWS Lambda
by awslabs
Wrap existing stdio‑based MCP servers to run as AWS Lambda functions, exposing them via HTTPS, API Gateway, Bedrock AgentCore, Lambda function URLs or the Lambda Invoke API.
EdgeOne Pages MCP
by TencentEdgeOne
Deploy HTML, folders, or zip archives to EdgeOne Pages and instantly obtain a public URL for fast edge delivery.
Volcengine MCP Servers
by volcengine
A comprehensive collection of Model Context Protocol (MCP) servers that expose Volcengine cloud resources and third‑party services through natural‑language interfaces, enabling AI‑driven operations across compute, storage, databases, networking, security, and developer utilities.
AWS MCP Server
by rishikavikondala
Provides Model Context Protocol tools for performing AWS S3 and DynamoDB operations, with automatic logging and audit access via the `audit://aws-operations` endpoint.
Mcp Confluent
by confluentinc
Enables AI assistants to manage Confluent Cloud resources such as Kafka topics, connectors, and Flink SQL statements through natural‑language interactions.
Alibaba Cloud Ops MCP Server
Officialby aliyun
Enables AI assistants to operate Alibaba Cloud resources such as ECS, Cloud Monitor, OOS and other services through seamless integration with Alibaba Cloud APIs via the Model Context Protocol.