MCP.so
登录
M

MIKU_Search

@GobinFan

代码仓库

关于 MIKU_Search

- 基于 AI 的智能搜索服务，支持多语言（简体中文、繁体中文、英文、日文、韩文、俄文、阿拉伯文）。

#ai#search-engine#multi-agent

基本信息

分类

搜索

传输方式

stdio

发布者

GobinFan

提交者

Gobin Fan

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "miku_search": {
      "command": "uv",
      "args": [
        "--directory",
        "/MIKU_MCP_SERVER",
        "run",
        "main.py"
      ],
      "env": {
        "MIKU_API_KEY": "<YOUR_MIKU_API_KEY>"
      }
    },
    "miku_search_sse": {
      "url": "https://hellomiku.cn/sse?apikey=<YOUR_MIKU_API_KEY>"
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is MIKU_Search?

MIKU_Search is an MCP server that integrates with the MIKU AI search ecosystem. It provides intelligent search request processing and distribution, supporting multiple languages and search modes for developers to integrate and extend. It is designed for users seeking an AI-powered, personalized, and real-time search experience via the MIKU platform.

How to use MIKU_Search?

  1. Obtain an API KEY from the MIKU Search open platform.
  2. Set up a Python virtual environment, install dependencies from requirements.txt, and configure the API key in a .env file as MIKU_API_KEY.
  3. Run the server with uv run main.py.
  4. Configure MCP clients (e.g., Cursor, Cline) using either the stdio transport (with uv run main.py) or the SSE transport (with a URL containing the API key).

Key features of MIKU_Search

  • AI-powered intelligent search with multi-language support (7 languages).
  • Two search modes: simple_search and deep_search.
  • Authentication via API KEY for data security.
  • Returns structured results including sources, related questions, and summary.
  • Integrates with MCP-compatible clients using stdio or SSE transports.

Use cases of MIKU_Search

  • Integrating an AI search engine into AI assistants or development tools.
  • Performing deep, context-aware searches for complex queries.
  • Retrieving multi-source search results with automatic summarization.
  • Enabling real-time, personalized search in multi-language applications.

FAQ from MIKU_Search

How do I get an API key for MIKU_Search?

Visit https://platform.hellomiku.cn/platform/home/ to register and obtain an API key.

What search modes are available?

Two modes are supported: simple_search for quick queries and deep_search for more thorough, context-rich results.

What languages does MIKU_Search support?

It supports Simplified Chinese, Traditional Chinese, English, Japanese, Korean, Russian, and Arabic.

How is authentication handled?

Authentication is performed via an API key (MIKU_API_KEY) passed either through environment variables or as a query parameter in SSE URLs.

Where can I find more documentation?

Refer to the MIKU Search open platform (https://platform.hellomiku.cn/platform/home/) and product docs (https://hyperspace.feishu.cn/wiki/BokYwwjreiXg8pkffDhcwXnW).

评论

搜索 分类下的更多 MCP 服务器

🚀 OneSearch MCP Server: Web Search & Crawl & Scraper & Extract

yokingma

🚀 OneSearch MCP Server: Web Search & Scraper & Extract, Support agent-browser, SearXNG, Tavily, DuckDuckGo, Bing, etc.

127

Serper Search and Scrape MCP Server

marcopesani

Serper MCP Server supporting search and webpage scraping

158

MCP SearXNG Enhanced Server

OvertliDS

Enhanced MCP server for SearXNG: category-aware web-search, web-scraping, and date/time retrieval.

54

duckduckgo-search MCP Server

zhsama

84

Google Search Tool

web-agent-master

A Playwright-based Node.js tool that bypasses search engine anti-scraping mechanisms to execute Google searches. Local alternative to SERP APIs with MCP server integration.

611