Alibaba Tablestore

v1.0.0โ€ขDatabasesโ€ขstable

โ˜• ๐Ÿ โ˜๏ธ - MCP service for Tablestore, features include adding documents, semantic search for documents based on vectors and scalars, RAG-friendly, and serverless.

alibabacloud-tablestoremcpai-integration
Share:
156
Stars
0
Downloads
0
Weekly
0/5

What is Alibaba Tablestore?

Alibaba Tablestore is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to โ˜• ๐Ÿ โ˜๏ธ - mcp service for tablestore, features include adding documents, semantic search for documents based on vectors and scalars, rag-friendly, and serverless.

โ˜• ๐Ÿ โ˜๏ธ - MCP service for Tablestore, features include adding documents, semantic search for documents based on vectors and scalars, RAG-friendly, and serverless.

This server falls under the Databases category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • โ˜• ๐Ÿ โ˜๏ธ - MCP service for Tablestore, features include adding

Use Cases

โ˜• ๐Ÿ โ˜๏ธ - MCP service for Tablestore, features include adding documents, semantic
aliyun

Maintainer

LicenseApache-2.0
Languagejava
Versionv1.0.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx alibabacloud-tablestore

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Alibaba Tablestore

The Alibaba Cloud Tablestore MCP Server connects AI assistants to Alibaba Cloud's Tablestore (OTS) service, enabling document storage and hybrid vector-scalar semantic search for RAG (Retrieval-Augmented Generation) workflows. It uses built-in embedding models (defaulting to BAAI/bge-base-zh-v1.5 via HuggingFace or Alibaba's DashScope API) to automatically vectorize documents at write time and query time, then performs hybrid vector and full-text retrieval against Tablestore's multi-dimensional index. This makes it ideal for building private knowledge base Q&A systems on a serverless, pay-as-you-go infrastructure.

Prerequisites

  • Python 3.10 or later installed, plus the uv package manager
  • An Alibaba Cloud account with a Tablestore (OTS) instance created
  • Tablestore credentials: TABLESTORE_INSTANCE_NAME, TABLESTORE_ENDPOINT, TABLESTORE_ACCESS_KEY_ID, TABLESTORE_ACCESS_KEY_SECRET
  • Optional: DashScope API key if using cloud-based embeddings instead of local HuggingFace models
  • An MCP-compatible client such as Claude Desktop or Cherry Studio
1

Create a Tablestore instance on Alibaba Cloud

Log in to the Alibaba Cloud Console, navigate to Tablestore (OTS), and create a new instance. Note the instance name and endpoint URL. The MCP server will automatically create the required tables and indexes on first run.

2

Install uv and set up the Python environment

Install the uv package manager and create a Python 3.10+ virtual environment in the cloned repository.

pip3 install uv
git clone https://github.com/aliyun/alibabacloud-tablestore-mcp-server.git
cd alibabacloud-tablestore-mcp-server/tablestore-python-mcp-server
uv python install 3.12
uv venv --python 3.12
3

Set required environment variables

Export your Alibaba Cloud credentials. These four variables are required; all others have defaults.

export TABLESTORE_ACCESS_KEY_ID=your_access_key_id
export TABLESTORE_ACCESS_KEY_SECRET=your_access_key_secret
export TABLESTORE_ENDPOINT=https://your-instance.cn-hangzhou.ots.aliyuncs.com
export TABLESTORE_INSTANCE_NAME=your_instance_name
4

Run the MCP server

Start the server in SSE mode (default) for shared access, or add --transport stdio for direct stdio integration.

# Optional: speed up HuggingFace model downloads in China
export HF_ENDPOINT=http://hf-mirror.com

uv run tablestore-mcp-server
5

Configure your MCP client to connect

For SSE mode the server runs on port 8001 by default. Configure your MCP client to connect to the SSE endpoint, or use stdio transport for direct client integration.

Alibaba Tablestore Examples

Client configuration

Claude Desktop configuration using stdio transport. Set all required Tablestore environment variables in the env block. Adjust the path to your Python binary.

{
  "mcpServers": {
    "tablestore": {
      "command": "/path/to/alibabacloud-tablestore-mcp-server/tablestore-python-mcp-server/.venv/bin/python",
      "args": ["-m", "tablestore_mcp_server"],
      "env": {
        "TABLESTORE_ACCESS_KEY_ID": "your_access_key_id",
        "TABLESTORE_ACCESS_KEY_SECRET": "your_access_key_secret",
        "TABLESTORE_ENDPOINT": "https://your-instance.cn-hangzhou.ots.aliyuncs.com",
        "TABLESTORE_INSTANCE_NAME": "your_instance_name"
      }
    }
  }
}

Prompts to try

These prompts demonstrate storing documents and performing semantic search through the two core tools.

- "Store this article in the knowledge base: [paste article text]"
- "Search the knowledge base for documents about machine learning optimization"
- "Find the most relevant documents about Python async programming"
- "Add these meeting notes to Tablestore and then search for action items related to Q3"
- "What documents do we have about database indexing strategies?"

Troubleshooting Alibaba Tablestore

Authentication error: 'OTSAuthFailed' or invalid credentials

Verify that TABLESTORE_ACCESS_KEY_ID and TABLESTORE_ACCESS_KEY_SECRET are correct RAM user credentials with Tablestore permissions. The TABLESTORE_ENDPOINT must match your instance's region endpoint exactly as shown in the Alibaba Cloud Console.

Embedding model download is slow or fails

The default HuggingFace model (BAAI/bge-base-zh-v1.5) is downloaded on first run. Set HF_ENDPOINT=http://hf-mirror.com to use a Chinese mirror. Alternatively, switch to DashScope cloud embeddings by setting EMBEDDING_PROVIDER_TYPE=dashscope and DASHSCOPE_API_KEY=your_key.

Table or index creation fails on first run

Ensure your RAM user has the necessary Tablestore permissions: ots:CreateTable, ots:CreateSearchIndex, ots:PutRow, ots:Search. You can grant these through the Alibaba Cloud RAM console policy editor.

Frequently Asked Questions about Alibaba Tablestore

What is Alibaba Tablestore?

Alibaba Tablestore is a Model Context Protocol (MCP) server that โ˜• ๐Ÿ โ˜๏ธ - mcp service for tablestore, features include adding documents, semantic search for documents based on vectors and scalars, rag-friendly, and serverless. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Alibaba Tablestore?

Follow the installation instructions on the Alibaba Tablestore GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Alibaba Tablestore?

Alibaba Tablestore works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Alibaba Tablestore free to use?

Yes, Alibaba Tablestore is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.

Browse More Databases MCP Servers

Explore all databases servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "alibabacloud-tablestore": { "command": "npx", "args": ["-y", "alibabacloud-tablestore"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide โ†’

Ready to use Alibaba Tablestore?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides