NotionMCP Light
NotionMCP Light is a Model Context Protocol (MCP) server that uses the Notion API to sync Markdown files with Notion pages.
overview
This project was developed to address the inefficiencies of Notion's official Model Context Protocol (MCP) server (which reads and writes Markdown in blocks and consumes LLM tokens). It provides an unofficial MCP server that does not consume tokens and can sync Markdown files directly with Notion's pages/database via API.
function
- Markdown → Notion
- Recognize H1 as page title
- Create Markdown content as a Notion page or a database page
- Database ID can be specified
- Using the Notion API directly (no tokens)
- Notion → Markdown
- Converts a given page or a page in a database to Markdown format.
- Output title as H1
- Convert block structure to Markdown
- Save to File
- MCP server support
- Model Context Protocol (MCP) compliant
- Provides endpoints that can be called from AI tools such as Cursor and Cline
- JSON-RPC over stdio based
install
Installing dependencies
Setting up Notion API Token
- Create an account on Notion Developers and get an API token.
- You can set it in an environment variable or create a
.env
file and set the token.
How to use
Starting the MCP Server
When using UV (recommended)
Or, if you want to specify the token directly:
If you use regular Python
Or, if you want to specify the token directly:
Setting in Cline/Cursor
To use NotionMCP Light with AI tools such as Cline/Cursor, add the following settings to your mcp_settings.json
file:
Replace /path/to/notion-mcp-light
with the installation directory of NotionMCP Light.
How to use MCP Tools
NotionMCP Light offers the following MCP tools:
uploadMarkdown
Upload your Markdown file and create it as a Notion page.
downloadMarkdown
Download your Notion page and save it as a Markdown file.
license
This project is released under the MIT License, see the LICENSE file for details.
This server cannot be installed
A Model Context Protocol server that synchronizes Markdown files with Notion pages without using LLM tokens, enabling direct interaction with Notion databases and pages through API calls.