Wolai MCP Server
Enables AI agents to interact with Wolai knowledge bases to read, write, and search pages, as well as manage databases and file uploads. It supports comprehensive content operations including block management and hierarchy navigation via the Model Context Protocol.
README
Wolai MCP Server 🐺
English | 中文
Connect AI agents to your Wolai knowledge base via the Model Context Protocol (MCP).
Read, write, search, and navigate your Wolai pages — all from Claude, Cursor, OpenCode, or any MCP-compatible AI agent.
✨ Features
| Category | Tools | Description |
|---|---|---|
| 📖 Read | get_block, get_block_children, get_breadcrumbs |
Read pages, list children, navigate hierarchy |
| ✏️ Write | create_block, update_block, delete_block |
Create, update, and delete blocks |
| 📊 Database | get_database, create_database_row, query_database_rows |
Query and modify database tables |
| 🔍 Search | search_pages |
Fuzzy title search across page tree |
| 📎 Upload | upload_file |
Upload file attachments to pages |
| 🔑 Token | create_token, refresh_token |
Manage API authentication tokens |
| ⚙️ Config | get_config, update_config |
Runtime configuration management |
16 tools total — covering read, write, database, search, upload, token, and configuration.
🚀 Quick Start
Install
npm install -g wolai-mcp
# or
npx wolai-mcp
Get Credentials
- Go to Wolai Developer Console
- Create an application → get App ID and App Secret
- Find the Root Page ID from your Wolai page URL (the part after
wolai.com/)
Interactive Setup
npx wolai-mcp init
Follow the prompts to enter your credentials. They will be saved to ~/.wolai-mcp/config.json.
Environment Variables
Alternatively, set environment variables:
export WOLAI_APP_ID="your_app_id"
export WOLAI_APP_SECRET="your_app_secret"
export WOLAI_ROOT_ID="your_root_page_id" # optional, for search
📋 Platform Configuration
OpenCode
Add to your opencode.json:
{
"mcp": {
"wolai-mcp": {
"command": "npx",
"args": ["wolai-mcp"],
"env": {
"WOLAI_APP_ID": "your_app_id",
"WOLAI_APP_SECRET": "your_app_secret",
"WOLAI_ROOT_ID": "your_root_page_id"
}
}
}
}
Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"wolai-mcp": {
"command": "npx",
"args": ["wolai-mcp"],
"env": {
"WOLAI_APP_ID": "your_app_id",
"WOLAI_APP_SECRET": "your_app_secret",
"WOLAI_ROOT_ID": "your_root_page_id"
}
}
}
}
Cursor / CherryStudio / Other MCP Clients
{
"wolai-mcp": {
"command": "npx",
"args": ["wolai-mcp"],
"env": {
"WOLAI_APP_ID": "your_app_id",
"WOLAI_APP_SECRET": "your_app_secret",
"WOLAI_ROOT_ID": "your_root_page_id"
}
}
}
💡 Usage Examples
Once configured, ask your AI agent:
- "读取我 Wolai 知识库的首页内容"
- "搜索标题包含'项目计划'的页面"
- "在指定页面下创建一个新页面叫'会议纪要'"
- "往指定页面添加一段代码"
- "查询我的数据表格内容"
- "显示当前 Wolai 配置状态"
🔧 Tool Reference
Token Management
create_token
Create a new API token.
app_id(string): Wolai Application IDapp_secret(string): Wolai Application Secret
refresh_token
Refresh the current API token.
Block Operations
get_block
Get details of a specific block or page.
id(string): Block ID or Page ID
get_block_children
Get child blocks of a given block or page.
id(string): Parent block IDcursor(string, optional): Pagination cursorpage_size(number, optional): Results per page (default: 100)
create_block
Create a new block under a parent.
parent_id(string): Parent block or page IDtype(string): Block type (text, heading, bullet, code, etc.)content(string, optional): Block content texttext_alignment(string, optional): left, center, or rightlevel(number, optional): Heading level 1-3language(string, optional): Code language for code blocks
update_block
Update an existing block.
id(string): Block IDcontent(string, optional): New contenttext_alignment(string, optional): New alignmentlevel(number, optional): New heading level
delete_block
Delete a block.
id(string): Block ID
get_breadcrumbs
Get the breadcrumb path for a block.
id(string): Block ID
Database Operations
get_database
Get database structure and data.
id(string): Database ID
create_database_row
Add rows to a database.
database_id(string): Database IDcells(string): JSON string of cell data
query_database_rows
Query database rows with pagination.
database_id(string): Database IDcursor(string, optional): Pagination cursorpage_size(number, optional): Results per page
Search
search_pages
Search pages by title keyword.
query(string): Search keywordroot_id(string, optional): Root page to start search from
Upload
upload_file
Upload a file attachment.
parent_id(string): Parent block or page IDfile_path(string): Local file path
Configuration
get_config
Get current configuration (hides secret).
update_config
Update configuration.
app_id(string, optional): New App IDapp_secret(string, optional): New App Secretroot_id(string, optional): New root page ID
📄 License
MIT License — see LICENSE for details.
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。