Notion
Manage and interact with your entire Notion workspace.
Browse your entire Notion workspace, not just one database
Markdown based Notion navigating MCP with just a single ROOT_PAGE variable, eliminating the need for a token.
- Notion MCP Server:
notion-texonom - Notion pages are converted into
text/markdownmimeType notes. - Search and retrieve relevant pages based on graph distance, considering parent-child and reference relationships.
A Model Context Protocol (MCP) server for managing and interacting with Notion-based notes. This TypeScript-based server demonstrates MCP concepts by integrating resources, tools, and prompts to interact with Notion pages efficiently.
Features
Resources
- Access Notes: List and retrieve Notion pages as
note://URIs with UUID slugs. - Metadata: Each resource includes a title, description, and content in Markdown format.
- Mime Types: Content is accessible in
text/markdownformat.
Tools
- Search Notes: Use the
search_notestool to search for Notion pages using a query string.- Input: Query text to filter relevant pages.
- Output: Markdown content of matching notes.
Prompts
- Summarize Notes: Generate summaries for individual Notion pages.
- Available Prompts:
summarize_note: Summarize a specific note by URI.suggest_refactor: Propose structural improvements.suggest_fix: Identify potential fixes for note content.suggest_enhance: Recommend enhancements to improve the note.
- Input: Notion page URI.
- Output: Structured messages for summarization and enhancement.
- Available Prompts:
Development
Setup
Install dependencies:
pnpm install
Build the project:
pnpm build
For development with auto-rebuild:
pnpm watch
Configuration
To configure the server with Notion:
- Set environment variables:
ROOT_PAGE: The root page ID of your Notion workspace.
Installation for Claude Desktop
To use this server with Claude Desktop, add the configuration:
- MacOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%/Claude/claude_desktop_config.json
Example configuration:
{
"mcpServers": {
"notion-texonom": {
"command": "node",
"args": [
"/path/to/mcp/build/index.js"
],
"env": {
"ROOT_PAGE": "$SOME_UUID"
}
}
}
}
Debugging
For troubleshooting and debugging the MCP server, use the MCP Inspector. To start the Inspector, run:
pnpm inspector
The Inspector provides a browser-based interface for inspecting stdio-based server communication.
Key Technologies
- Notion Integration: Powered by
@texonom/nclientand@texonom/cli. - MCP SDK: Implements
@modelcontextprotocol/sdkfor server operations.
Remote Deployment
The server now uses SSEServerTransport for remote communication, enabling shared usage of the server. Ensure that the necessary dependencies are installed and the server is configured correctly for remote deployment.
Usage Instructions
To run the server with SSEServerTransport, use the following command:
npx -y supergateway --port 8000 --stdio "npx -y @modelcontextprotocol/server-filesystem /some/folder"
Make sure to replace /some/folder with the appropriate path to your folder.
関連サーバー
Markdown to any text
Convert markdown to any text format you want
Google Calendar
Integrates with Google Calendar to manage events and generate calendar insights.
ONLYOFFICE DocSpace
Interact with ONLYOFFICE DocSpace for document collaboration and management.
Taskade
Connect to the Taskade platform via MCP. Access tasks, projects, workflows, and AI agents in real-time through a unified workspace and API.
NPX-MCP
A TypeScript server for automation and integration, featuring Microsoft OAuth, browser control, and basic utilities.
AI2Fin | Intelligent Financial Management
Tools for expense tracking, tax deductions, bill management, and financial analysis.
Integrator (legacy)
Use Integrator scenarios as tools for AI assistants.
JSON Canvas MCP Server
A server for creating, modifying, validating, and exporting JSON Canvas files, a format for infinite canvas data.
Jira MCP Server by CData
A read-only MCP server for Jira, enabling LLMs to query live Jira data using the CData JDBC Driver.
Attio
Connect your Attio workspace to large language models.