Alibaba Cloud MCP Server
Provides tools for managing Alibaba Cloud resources, including ECS, VPC, RDS, RAM, ACK, SLB, and a universal API invoker for all services.
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., "@Alibaba Cloud MCP Server list my ECS instances"
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.
Alibaba Cloud MCP Server
A comprehensive, high-performance Model Context Protocol (MCP) server for managing Alibaba Cloud resources.
This server enables AI assistants (like Claude, Cursor, and others) to seamlessly interact with Alibaba Cloud infrastructure using natural language. It features a Hybrid Architecture:
Explicit Service Tools: Pre-configured, typed handlers for the most common services (ECS, VPC, RDS, RAM, ACK, SLB) to ensure rapid, error-free AI interactions.
Universal API Invoker: A dynamic tool built on
@alicloud/pop-corethat can invoke any of the 10,000+ API endpoints across all Alibaba Cloud services, giving you true 100% cloud management coverage.
Features
🚀 Zero-Config Universal Management: Dynamically call any Alibaba Cloud RPC/ROA API.
📦 Explicit Typed Tools:
ECS: Manage Compute Instances (
ecs_list_instances,ecs_start_instance)VPC: Manage Networking (
vpc_list,vpc_create)RDS: Manage Databases (
rds_list_instances)RAM: Manage IAM & Security (
ram_list_users)ACK: Manage Kubernetes Clusters (
ack_list_clusters)SLB: Manage Load Balancers (
slb_list)
🐳 Docker Ready: Run safely in an isolated container.
🛡️ Secure: Uses Zod for strict parameter validation and environment variables for credentials.
Related MCP server: MCP Cloud Services Server
Prerequisites
Node.js 18+ (if running locally)
Docker (optional, for containerized execution)
Alibaba Cloud Access Key ID and Secret with appropriate IAM permissions.
Setup & Installation
Option 1: Running Locally
Clone the repository and install dependencies:
npm installBuild the TypeScript source:
npm run buildCreate a
.envfile in the root directory:ALIBABA_CLOUD_ACCESS_KEY_ID="your_access_key" ALIBABA_CLOUD_ACCESS_KEY_SECRET="your_access_secret" ALIBABA_CLOUD_REGION_ID="cn-hangzhou"
Option 2: Running with Docker
Build the Docker image:
docker build -t alibaba-cloud-mcp .
Configuring Your MCP Client
Claude Desktop
To integrate this server with Claude Desktop, add the following to your claude_desktop_config.json (usually located at %APPDATA%\Claude\claude_desktop_config.json on Windows or ~/Library/Application Support/Claude/claude_desktop_config.json on Mac):
If using Node.js locally:
{
"mcpServers": {
"alibaba-cloud": {
"command": "node",
"args": ["/absolute/path/to/Alibaba_cloud_MCP_server/dist/index.js"],
"env": {
"ALIBABA_CLOUD_ACCESS_KEY_ID": "your_access_key",
"ALIBABA_CLOUD_ACCESS_KEY_SECRET": "your_access_secret",
"ALIBABA_CLOUD_REGION_ID": "cn-hangzhou"
}
}
}
}If using Docker:
{
"mcpServers": {
"alibaba-cloud": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e", "ALIBABA_CLOUD_ACCESS_KEY_ID",
"-e", "ALIBABA_CLOUD_ACCESS_KEY_SECRET",
"-e", "ALIBABA_CLOUD_REGION_ID",
"alibaba-cloud-mcp"
],
"env": {
"ALIBABA_CLOUD_ACCESS_KEY_ID": "your_access_key",
"ALIBABA_CLOUD_ACCESS_KEY_SECRET": "your_access_secret",
"ALIBABA_CLOUD_REGION_ID": "cn-hangzhou"
}
}
}
}Note: Restart Claude Desktop after updating the configuration.
Security Warning
⚠️ IMPORTANT: This MCP server is incredibly powerful, especially due to the
aliyun_invoke_apitool which can manage any resource. Always adhere to the Principle of Least Privilege. Ensure that the IAM/RAM user associated with your Access Keys only has the minimum permissions necessary for the tasks you intend the AI to perform.
Architecture
Built using:
@modelcontextprotocol/sdk@alicloud/pop-corezodfor validationTypeScript
Available Tools
9 toolsack_list_clustersA
List Container Service for Kubernetes (ACK) clusters.
Inputs: no named parameters.
aliyun_invoke_apiA
Invoke ANY Alibaba Cloud OpenAPI endpoint dynamically. This tool allows you to manage ALL of Alibaba Cloud. Use the OpenAPI documentation to find the correct endpoint, apiVersion, and action. Note: RPC requests are most common.
Inputs: endpoint (required), apiVersion (required), action (required), parameters, +1 more.
ecs_list_instancesB
List Alibaba Cloud ECS instances in the current region.
Inputs: regionId, pageSize.
ecs_start_instanceB
Start an Alibaba Cloud ECS instance.
Inputs: instanceId (required).
ram_list_usersB
List IAM/RAM users.
Inputs: no named parameters.
rds_list_instancesC
List RDS instances.
Inputs: regionId.
slb_listC
List Server Load Balancers (SLB).
Inputs: regionId.
vpc_createC
Create a Virtual Private Cloud (VPC).
Inputs: cidrBlock (required), regionId, vpcName.
vpc_listC
List Virtual Private Clouds (VPCs).
Inputs: regionId, pageSize.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 9 tools
Most tools target distinct services (ECS, VPC, RDS, etc.), but the generic 'aliyun_invoke_api' can overlap with any specific tool, causing potential confusion. Within services, actions like 'ecs_list_instances' and 'ecs_start_instance' are clearly distinct.
Tools generally follow a service_verb_noun pattern (e.g., ecs_list_instances), but some omit the resource noun (slb_list, vpc_create) and 'aliyun_invoke_api' uses a different format, leading to inconsistency.
9 tools for a cloud provider is reasonable; the generic API invocation tool extends capability. However, the scope is broad, and a few more targeted tools could improve coverage without being excessive.
The tool set is very incomplete for most services—only list operations for RAM, RDS, SLB; no update, delete, or advanced actions. The generic API tool is a workaround but indicates significant gaps in native coverage.
Related MCP Connectors
Unified API to query AWS, GCP, Azure and generate Terraform/CLI execution kits for AI agents.
Deploy, monitor, and manage your OpenClaw AI assistants via natural language.
Provides capabilities that let LLM agents perform a range of infrastructure management tasks.
214 tools for AI agents over one MCP endpoint: image, video, audio, 3D, data, storage, mail, GPUs.
Related MCP Servers
AlicenseNot gradedqualityDmaintenanceEnables AI assistants to manage Supabase projects deployed on Alibaba Cloud through natural language commands. Supports project creation/deletion, database querying, security configuration, and table management operations.8Apache 2.0- AlicenseCqualityDmaintenanceEnables AI assistants to manage multi-cloud resources (AWS, Azure, GCP) including resource operations, cost analysis, monitoring metrics, and security compliance checks through natural language commands.2613 npm2MIT
- AlicenseCqualityAmaintenanceEnables AI assistants to interact with Alibaba Cloud Yunxiao DevOps platform for managing projects, code repositories, work items, pipelines, deployments, and testing workflows through comprehensive organization, development, and delivery tools.813,142 npm185Apache 2.0
- AlicenseNot gradedqualityDmaintenanceProvides AI agents with natural language control over AWS, Azure, GCP, and Alibaba Cloud infrastructure through dynamic API discovery and execution. Supports 51,900+ cloud operations and includes OpenTofu integration for complete infrastructure lifecycle management.3MIT