Alibabacloud Opensearch MCP Server

@aliyun

代码仓库

基本信息

分类

数据库

传输方式

stdio

作者

aliyun

提交者

bestCMC

关于 Alibabacloud Opensearch MCP Server

This MCP server equips AI Agents with tools to interact with OpenSearch through a standardized and extensible interface. See our product document for more information.

#ai #aliyun #opensearch

Alibabacloud Opensearch MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "aisearch-mcp-server": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/aisearch-mcp-server",
        "run",
        "aisearch-mcp-server"
      ],
      "env": {
        "AISEARCH_API_KEY": "<AISEARCH_API_KEY>",
        "AISEARCH_ENDPOINT": "<AISEARCH_ENDPOINT>"
      }
    },
    "opensearch-vector-mcp-server": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/opensearch-vector-mcp-server",
        "run",
        "opensearch-vector-mcp-server"
      ],
      "env": {
        "OPENSEARCH_VECTOR_ENDPOINT": "http://ha-cn-***.public.ha.aliyuncs.com",
        "OPENSEARCH_VECTOR_USERNAME": "<username>",
        "OPENSEARCH_VECTOR_PASSWORD": "<password>",
        "OPENSEARCH_VECTOR_INSTANCE_ID": "ha-cn-***",
        "OPENSEARCH_VECTOR_INDEX_NAME": "<Optional: index in vector table>",
        "AISEARCH_API_KEY": "<Optional: AISEARCH_API_KEY for embedding>",
        "AISEARCH_ENDPOINT": "<Optional: AISEARCH_ENDPOINT for embedding>"
      }
    }
  }
}

Alibabacloud Opensearch MCP Server 工具

001 provides a text vectorization service that supports more than 40 languages. The input text can be up to 300 tokens in length, and the dimension of the generated vectors is 1,536.

002 provides a text vectorization service that supports more than 100 languages. The input text can be up to 8,192 tokens in length, and the dimension of the generated vectors is 1,024.

001 provides a text vectorization service for Chinese text. The input text can be up to 1,024 tokens in length, and the dimension of the generated vectors is 768.

001 provides a text vectorization service for English text. The input text can be up to 512 tokens in length, and the dimension of the generated vectors is 768.

概览

What is Alibabacloud Opensearch Mcp Server?

Alibabacloud Opensearch Mcp Server serves as a universal interface between AI Agents and the OpenSearch AI Search Platform and OpenSearch Vector service from Alibaba Cloud. It is designed for developers who want to integrate AI agents with Alibaba Cloud's search, vector, and document processing capabilities.

How to use Alibabacloud Opensearch Mcp Server?

Clone the repository from GitHub, then add the appropriate JSON configuration to your MCP client. For the AISearch service, set AISEARCH_API_KEY and AISEARCH_ENDPOINT environment variables. For the OpenSearch Vector service, set OPENSEARCH_VECTOR_ENDPOINT, username, password, instance ID, and optionally the embedding API key and endpoint.

Key features of Alibabacloud Opensearch Mcp Server

Document parsing with logical structure extraction (AISearch)
Image content analysis using multimodal LLMs (AISearch)
Text splitting and embedding generation in multiple languages (AISearch)
Hybrid search combining dense and sparse vectors (OpenSearch Vector)
Query analysis: user intent, question expansion, and NL-to-SQL (AISearch)
Internet search to supplement private knowledge bases (AISearch)

Use cases of Alibabacloud Opensearch Mcp Server

Build retrieval-augmented generation (RAG) applications by parsing and vectorizing documents.
Perform conversational search with intent understanding and SQL generation.
Enhance search accuracy with re-ranking and hybrid dense/sparse retrieval.
Combine private data with real-time internet information for richer answers.

常见问题

What runtime dependencies are required?

The server requires Python and the `uv` package manager. Environment variables must be configured for authentication and endpoint selection.

How do I configure authentication for the two services?

For AISearch, set `AISEARCH_API_KEY` and `AISEARCH_ENDPOINT`. For OpenSearch Vector, set `OPENSEARCH_VECTOR_ENDPOINT`, `OPENSEARCH_VECTOR_USERNAME`, `OPENSEARCH_VECTOR_PASSWORD`, and `OPENSEARCH_VECTOR_INSTANCE_ID`.

What are the token limits for text embedding models?

Ops-text-embedding-001 supports up to 300 tokens; ops-text-embedding-002 up to 8,192; ops-text-embedding-zh-001 up to 1,024; ops-text-embedding-en-001 up to 512 tokens.

How does the server handle sparse and dense vector search?

The OpenSearch Vector component provides tools for simple vector search, multi-query search, and combined search with a dense vector plus a sparse vector.