Skip to main content

基于 Everything 搜索引擎的 MCP (Model Context Protocol) 服务器,让 AI 助手能够搜索本地文件。

Project description

Everything MCP

基于 Everything 搜索引擎的 MCP (Model Context Protocol) 服务器。让你的 AI 助手(如 Claude、Kiro 等)能够快速搜索本地文件。

前置要求

  • Windows 系统(Everything 仅支持 Windows)
  • Everything 已安装并正在运行
  • Python >= 3.12

安装

pip install everything-search-mcp

或使用 uv

uv pip install everything-search-mcp

MCP 配置

Claude Desktop

编辑 Claude Desktop 配置文件(通常位于 %APPDATA%\Claude\claude_desktop_config.json):

{
  "mcpServers": {
    "everything": {
      "command": "everything-search-mcp",
      "disabled": false
    }
  }
}

如果使用 uvx 运行(无需预先安装):

{
  "mcpServers": {
    "everything": {
      "command": "uvx",
      "args": ["everything-search-mcp"],
      "disabled": false
    }
  }
}

Kiro

在项目根目录或用户目录下创建 .kiro/settings/mcp.json

{
  "mcpServers": {
    "everything": {
      "command": "everything-search-mcp",
      "disabled": false
    }
  }
}

使用 uvx

{
  "mcpServers": {
    "everything": {
      "command": "uvx",
      "args": ["everything-search-mcp"],
      "disabled": false
    }
  }
}

Cursor

编辑 ~/.cursor/mcp.json

{
  "mcpServers": {
    "everything": {
      "command": "everything-search-mcp",
      "disabled": false
    }
  }
}

OpenCode

编辑项目根目录下的 opencode.json(或全局配置):

{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "everything": {
      "type": "local",
      "command": ["everything-search-mcp"],
      "enabled": true
    }
  }
}

使用 uvx

{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "everything": {
      "type": "local",
      "command": ["uvx", "everything-search-mcp"],
      "enabled": true
    }
  }
}

VS Code (Copilot)

.vscode/settings.json 中添加:

{
  "mcp": {
    "servers": {
      "everything": {
        "command": "everything-search-mcp"
      }
    }
  }
}

提供的工具

get_current_date

获取当前日期(YYYY-MM-DD 格式),方便在日期筛选时使用。

get_everything_info

获取 Everything 的版本、数据库加载状态和管理员权限信息。

search

基础搜索工具,支持以下参数:

参数 类型 默认值 说明
query_string str 必填 搜索关键词
match_case bool False 区分大小写
match_path bool False 匹配完整路径
match_whole_word bool False 全词匹配
regex bool False 启用正则表达式
sort_type int 1 排序方式(见下方说明)
max_results int 10 最大返回数量
return_properties list 默认字段 返回的字段列表

complex_search

高级组合搜索,通过 filters 列表组合多种过滤条件:

  • keywords — 关键词搜索
  • file_filter — 按扩展名、大小范围、文件内容过滤
  • date_filter — 按创建/修改/访问日期范围过滤
  • size_filter — 按文件大小过滤(支持 kb/mb/gb 单位)
  • media_filter — 按媒体类型过滤(image/audio/video)
  • document_filter — 按文档类型过滤(office/pdf/text)

使用示例

配置完成后,你可以直接在 AI 对话中使用自然语言搜索文件:

"帮我找一下桌面上所有的 PDF 文件"

"搜索最近一周修改过的 Python 文件"

"找出所有大于 100MB 的视频文件"

"在 D 盘搜索包含 config 的 yaml 文件"

AI 助手会自动调用对应的 MCP 工具完成搜索。

排序类型参考

说明 说明
1 名称升序 2 名称降序
3 路径升序 4 路径降序
5 大小升序 6 大小降序
11 创建日期升序 12 创建日期降序
13 修改日期升序 14 修改日期降序

开发

git clone https://github.com/miloira/everything-search-mcp.git
cd everything-search-mcp
uv sync

本地测试运行:

uv run everything-search-mcp

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

everything_search_mcp-0.0.1.tar.gz (68.1 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

everything_search_mcp-0.0.1-py3-none-any.whl (8.1 kB view details)

Uploaded Python 3

File details

Details for the file everything_search_mcp-0.0.1.tar.gz.

File metadata

  • Download URL: everything_search_mcp-0.0.1.tar.gz
  • Upload date:
  • Size: 68.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.10.10 {"installer":{"name":"uv","version":"0.10.10","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":null,"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for everything_search_mcp-0.0.1.tar.gz
Algorithm Hash digest
SHA256 fdd73fb6bc12e1147c973e9133ae9d945808e984f193eec42cdbe75588c1b80b
MD5 0b2c5a8bd2612d51aa5350ef76ecf077
BLAKE2b-256 1087875b6a7e1af4ab097026521481bc5d295171ce2a342ae5efb9db3b4bcc5b

See more details on using hashes here.

File details

Details for the file everything_search_mcp-0.0.1-py3-none-any.whl.

File metadata

  • Download URL: everything_search_mcp-0.0.1-py3-none-any.whl
  • Upload date:
  • Size: 8.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.10.10 {"installer":{"name":"uv","version":"0.10.10","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":null,"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for everything_search_mcp-0.0.1-py3-none-any.whl
Algorithm Hash digest
SHA256 59550af9abbf4af6dbfa27109157b8fb2b3bbce556df7062d147b6609a925bfd
MD5 0e62165ab0b8fa6774de912e4fa4150e
BLAKE2b-256 5e69d30964c7a38aacc2fae78521a613975d3779496eb9ba04b7a538859337f1

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page