@growi/mcp-server

@growi/mcp-server

Connects AI models to GROWI wikis for search and retrieval, enabling context-aware responses from an organization's knowledge base.

Category
访问服务器

README

@growi/mcp-server

npm version License: MIT

A Model Context Protocol (MCP) server that connects AI models to GROWI wiki content. Enables LLMs to search and retrieve information from your organization's knowledge base for accurate, context-aware responses. Supports connections to multiple GROWI apps.

Key Features

  • 🔍 GROWI page search and retrieval
  • 📝 Page management
  • 🏷️ Tag management
  • 📋 Comment management
  • 🔗 Share link management

Supported GROWI Versions

  • GROWI v7.3.x or higher recommended
  • Some features are also available starting from GROWI v7.2.5 and later
  • GROWI API

Agent Skills

This repository also provides Agent Skills — reusable workflow definitions that AI coding agents can load to interact with GROWI more effectively.

Available Skills

  • growi-mcp-setup — Walks you through setting up the GROWI MCP server. After the skill is installed, it guides you from configuring UTCP Code-Mode through verifying the connection.
  • growi-smart-save — Save content to GROWI with intelligent path suggestions. The agent calls the suggest-path tool, presents destination candidates, and guides the user through page naming and visibility settings.

Quick Start (Recommended)

The fastest way to start using GROWI. Once you install the skill, the rest of the setup (connecting the MCP server, configuring UTCP Code-Mode, verifying connectivity) is guided by the AI agent.

1. Install the Skill

Install the skill for your agent.

Claude Desktop (Cowork)

  1. Go to Customize > Personal Plugins (click the + icon)
  2. Click Browse Plugins > select the Personal tab
  3. Click the + icon next to Local Upload
  4. Select Add marketplace from GitHub
  5. Enter the repository URL and click Sync:
https://github.com/growilabs/growi-mcp-server

Claude Code

Add this repository as a plugin marketplace, then install the plugin:

/plugin marketplace add growilabs/growi-mcp-server
/plugin install mcp-client-skills

Gemini CLI

Install as a Gemini CLI extension (includes both MCP tools and skills):

gemini extensions install https://github.com/growilabs/growi-mcp-server

Update with:

gemini extensions update growi-mcp-server

Skills.sh (Vercel)

Works with Claude Code, Gemini CLI, Cursor, Codex, GitHub Copilot, and many other agents:

npx skills add growilabs/growi-mcp-server

Update with:

npx skills update

Manual Installation

Download skills directly from the repository and place them in your agent's skills directory:

  1. Copy the desired skill directory from skills/ in this repository
  2. Place it in your agent's skills directory:
    • Claude Code: .claude/skills/<skill-name>/SKILL.md
    • Gemini CLI: .gemini/skills/<skill-name>/SKILL.md
    • Other agents: .agents/skills/<skill-name>/SKILL.md

2. Restart Your Agent

After installation, restart (or reload) your agent so the skill is recognized.

3. Ask the AI to Set Up

Tell your agent "set up GROWI", and the growi-mcp-setup skill will start and guide you from the MCP server connection settings through verifying connectivity.

[!NOTE] If you want to configure the MCP server directly without the skill, see Use the MCP Server Directly.

Use the MCP Server Directly

You can also register the MCP server directly with your agent instead of using the skill. Use this for a minimal setup, or as a fallback when skill-based setup is not available.

Supports simultaneous connections to multiple GROWI apps. Each app is configured using numbered environment variables.

Single App Configuration Example

{
  "mcpServers": {
    "growi": {
      "command": "npx",
      "args": ["@growi/mcp-server"],
      "env": {
        "GROWI_APP_NAME_1": "main",
        "GROWI_BASE_URL_1": "https://your-growi-instance.com",
        "GROWI_API_TOKEN_1": "your_growi_api_token"
      }
    }
  }
}

Multiple Apps Configuration Example

{
  "mcpServers": {
    "growi": {
      "command": "npx",
      "args": ["@growi/mcp-server"],
      "env": {
        "GROWI_DEFAULT_APP_NAME": "staging",

        "GROWI_APP_NAME_1": "production",
        "GROWI_BASE_URL_1": "https://wiki.example.com",
        "GROWI_API_TOKEN_1": "token_for_production",

        "GROWI_APP_NAME_2": "staging",
        "GROWI_BASE_URL_2": "https://wiki-staging.example.com",
        "GROWI_API_TOKEN_2": "token_for_staging",

        "GROWI_APP_NAME_3": "development",
        "GROWI_BASE_URL_3": "https://wiki-dev.example.com",
        "GROWI_API_TOKEN_3": "token_for_development"
      }
    }
  }
}

[!TIP] For skill-based setup (recommended), see Quick Start.

Available Tools (Features)

Page Management

  • searchPages - Search pages by keywords
  • createPage - Create a new page
  • updatePage - Update an existing page
  • deletePages - Delete pages (bulk operation supported)
  • duplicatePage - Duplicate a page (including child pages)
  • renamePage - Change page name and path
  • getPage - Get a page data
  • getPageInfo - Get detailed page information
  • getRecentPages - Get list of recently updated pages
  • getPageListingRoot - Get root page list
  • getPageListingChildren - Get child pages of specified page
  • pageListingInfo - Get summary information of page listings
  • publishPage / unpublishPage - Set page publish/unpublish status

Tag Management

  • getPageTag - Get tags of a page
  • updateTag - Update tags of a page
  • getTagList - Get list of tags
  • searchTags - Search tags

Comments & Discussions

  • getComments - Get comments of a page

Revision Management

  • listRevisions - Get page edit history
  • getRevision - Get details of a specific revision

Share Links

  • createShareLink - Create a share link
  • getShareLinks - Get share links of a page
  • deleteShareLinks - Delete share links
  • deleteShareLinkById - Delete a specific share link

User Information

  • getUserRecentPages - Get recent pages of a specific user

Configuration Options

Environment Variables

Variable Name Required Description Default Value
GROWI_APP_NAME_{N} GROWI app identifier name (N is an integer) -
GROWI_BASE_URL_{N} Base URL of GROWI instance (N is an integer) -
GROWI_API_TOKEN_{N} GROWI API access token (N is an integer) -
GROWI_DEFAULT_APP_NAME Default app name to use First configured app

Multiple Apps Configuration Notes

  • Use integer values (1, 2, 3...) for each app configuration (sequential numbering is not required)
  • Combination of GROWI_APP_NAME_N, GROWI_BASE_URL_N, and GROWI_API_TOKEN_N is required
  • App names, base URLs, and API tokens must each be unique
  • If GROWI_DEFAULT_APP_NAME is omitted, the first configured app becomes the default
  • The app specified in GROWI_DEFAULT_APP_NAME will be used as the default app when the LLM does not explicitly include an app name in the prompt

Developer Information

Requirements

  • Node.js 18 or higher
  • pnpm (recommended)
  • GROWI instance (for development and testing)

Getting Started

  1. Clone the repository
git clone https://github.com/growilabs/growi-mcp-server.git
cd growi-mcp-server
  1. Install dependencies
pnpm install
  1. Set up environment variables
cp .env.example .env.local
# Edit .env.local to enter GROWI connection information
  1. Start the development server
# Test with MCP CLI
pnpm dev:cli

# Develop with MCP Inspector
pnpm dev:inspect

Build and Test

# Build
pnpm build

# Lint
pnpm lint

# Run tests
pnpm test

# Run tests with coverage
pnpm test:coverage

# Run in production
pnpm start

MCP Server Configuration

  1. Build
pnpm build
  1. MCP Server Configuration (Single App)
{
  "mcpServers": {
    "growi": {
      "command": "node",
      "args": ["/Users/username/projects/growi-mcp-server/dist/index.js"],
      "env": {
        "GROWI_APP_NAME_1": "main",
        "GROWI_BASE_URL_1": "https://your-growi-instance.com",
        "GROWI_API_TOKEN_1": "your_growi_api_token"
      }
    }
  }
}
  1. MCP Server Configuration (Multiple Apps)
{
  "mcpServers": {
    "growi": {
      "command": "node",
      "args": ["/Users/username/projects/growi-mcp-server/dist/index.js"],
      "env": {
        "GROWI_DEFAULT_APP_NAME": "production",

        "GROWI_APP_NAME_1": "production",
        "GROWI_BASE_URL_1": "https://wiki.example.com",
        "GROWI_API_TOKEN_1": "production_token",

        "GROWI_APP_NAME_2": "staging",
        "GROWI_BASE_URL_2": "https://wiki-staging.example.com",
        "GROWI_API_TOKEN_2": "staging_token"
      }
    }
  }
}

[!NOTE] Set the absolute path to the built output in "args"

Troubleshooting

When unable to connect to GROWI

  1. Check connectivity
    curl -v http://app:3000/_api/v3/healthcheck
    
  2. If the app hostname cannot be resolved, check the devcontainer network and verify it includes growi_devcontainer_default
    • The .devcontainer/devcontainer.json file sets --network in runArgs, so rebuilding the container should apply this setting
    • To add manually, run the following:
      • Run docker network command on the docker host machine
      docker network connect growi_devcontainer_default growi-mcp-server-dev
      

Contributing

Contributions to the project are welcome!

How to Contribute

  1. Issue Reports: Bug reports and feature requests via GitHub Issues
  2. Pull Requests:
    • Fork and create a branch
    • Implement changes
    • Add tests (if applicable)
    • Create a pull request

Development Guidelines

License

This project is released under the MIT License.


Related Links


Notice

This MCP server is under development. APIs may change without notice. Please test thoroughly before using in production environments.

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