Nuclino MCP Server

Nuclino MCP Server

Provides access to Nuclino content through structured search and retrieval tools.

Category
访问服务器

README

Nuclino MCP Server

A Model Context Protocol (MCP) server that provides access to Nuclino content through structured search and retrieval tools.

Features

  • Search Nuclino content by team or workspace
  • Retrieve specific items by ID
  • Get team and workspace information
  • Find teams and workspaces by name
  • Rate limiting and retry logic for API reliability
  • Support for both HTTP and stdio transports

Transport Types

This server supports two transport types:

HTTP Transport (Default)

  • Best for web applications and remote access
  • Supports multiple concurrent sessions
  • Requires API key to be sent in headers for each request
  • Runs on port 3000 by default

Stdio Transport

  • Best for local development and CLI tools
  • Single session per server instance
  • Requires API key to be set as environment variable
  • Communicates over stdin/stdout

Installation

npm install

Configuration

Environment Variables

  • TRANSPORT_TYPE: Set to 'http' (default) or 'stdio'
  • PORT: Port for HTTP transport (default: 3000)
  • NUCLINO_API_KEY: Required for stdio transport only

HTTP Transport

  1. Start the server:
npm run dev  # Development
npm run start  # Production
  1. Connect to the server at http://localhost:3000/mcp

  2. Include your Nuclino API key in the nuclino-api-key header with each request.

Stdio Transport

  1. Set your API key:
export NUCLINO_API_KEY="your_api_key_here"
  1. Start the server:
npm run dev:stdio  # Development
npm run start:stdio  # Production

The server will communicate via stdin/stdout, making it suitable for direct integration with MCP clients.

API Key

You need a Nuclino API key to use this server. You can create one in your Nuclino profile page.

Usage

You will find an .mcp.json.example file in this repository to show how you can configure both versions of the MCP. Once installed, you can use the MCP to summarize all info it can find on Nuclino about certain topics. You can stear it by passing a workspace name to limit the search scope.

Available Tools

  • search_by_team: Search content within a specific team
  • search_by_workspace: Search content within a specific workspace
  • get_teams: Get all available teams
  • get_workspaces: Get all available workspaces
  • find_team_by_name: Find a team by name
  • find_workspace_by_name: Find a workspace by name
  • get_item: Get a specific item by ID

Development

Building

npm run build

Running in Development

# HTTP transport
npm run dev

# Stdio transport
npm run dev:stdio

Docker

# Build image
docker build -t nuclino-mcp-server .

# Run HTTP server
docker run -d -p 3000:3000 --name nuclino-mcp-server nuclino-mcp-server

# Run stdio server
docker run -e TRANSPORT_TYPE=stdio -e NUCLINO_API_KEY=your_key nuclino-mcp-server

Architecture

The server follows Clean Architecture principles:

  • Domain Layer: Entities and repository interfaces
  • Application Layer: Use cases and business logic
  • Infrastructure Layer: External integrations (Nuclino API, transports)
  • Presentation Layer: MCP server implementation

Transport Layer

The transport layer is abstracted to support multiple transport types:

  • ITransport: Common interface for all transports
  • HttpTransport: HTTP-based transport with session management
  • StdioTransport: Stdio-based transport for local usage
  • TransportFactory: Factory for creating transport instances

Error Handling

The server includes comprehensive error handling:

  • Rate limiting with exponential backoff
  • Retry logic for transient failures
  • Graceful shutdown handling
  • Detailed logging for debugging

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选