by makenotion
Provides a remote Model Context Protocol server for the Notion API, enabling OAuth‑based installation and optimized toolsets for AI agents with minimal token usage.
Notion MCP Server implements a Model Context Protocol (MCP) server that translates client requests into Notion API calls. It abstracts authentication, supports both STDIO and HTTP transports, and is packaged for easy deployment via npm or Docker.
ntn_…
).npx
and injects the token via the NOTION_TOKEN
environment variable (or OPENAPI_MCP_HEADERS
for custom headers).npx @notionhq/notion-mcp-server
npx @notionhq/notion-mcp-server --transport http --port 3000
npx
command with environment variables; works with popular AI assistants and editors.Q: Do I need to expose my Notion token to the LLM? A: The token is stored only in the server environment and never sent to the client. The LLM interacts with the server via JSON‑RPC, keeping the token private.
Q: Can I restrict the integration to read‑only? A: Yes. Set the integration capabilities to “Read content” only in the Notion developer portal.
Q: Which transport should I choose? A: Use STDIO for desktop AI assistants (Claude, Cursor). Use HTTP when building web‑based interfaces or when you need to expose the server over a network.
Q: How do I change the HTTP port or auth token?
A: Pass --port <number>
and --auth-token <token>
flags, or set the AUTH_TOKEN
environment variable.
Q: Is there a pre‑built Docker image?
A: Yes. Pull mcp/notion
from Docker Hub and run it with -e NOTION_TOKEN
(or OPENAPI_MCP_HEADERS
).
[!NOTE]
We’ve introduced Notion MCP, a remote MCP server with the following improvements:
- Easy installation via standard OAuth. No need to fiddle with JSON or API token anymore.
- Powerful tools tailored to AI agents. These tools are designed with optimized token consumption in mind.
Learn more and try it out here
This project implements an MCP server for the Notion API.
Go to https://www.notion.so/profile/integrations and create a new internal integration or select an existing one.
While we limit the scope of Notion API's exposed (for example, you will not be able to delete databases via MCP), there is a non-zero risk to workspace data by exposing it to LLMs. Security-conscious users may want to further configure the Integration's Capabilities.
For example, you can create a read-only integration token by giving only "Read content" access from the "Configuration" tab:
Ensure relevant pages and databases are connected to your integration.
To do this, visit the Access tab in your internal integration settings. Edit access and select the pages you'd like to use.
Alternatively, you can grant page access individually. You'll need to visit the target page, and click on the 3 dots, and select "Connect to integration".
Using npm:
Cursor & Claude:
Add the following to your .cursor/mcp.json
or claude_desktop_config.json
(MacOS: ~/Library/Application\ Support/Claude/claude_desktop_config.json
)
Option 1: Using NOTION_TOKEN (recommended)
{
"mcpServers": {
"notionApi": {
"command": "npx",
"args": ["-y", "@notionhq/notion-mcp-server"],
"env": {
"NOTION_TOKEN": "ntn_****"
}
}
}
}
Option 2: Using OPENAPI_MCP_HEADERS (for advanced use cases)
{
"mcpServers": {
"notionApi": {
"command": "npx",
"args": ["-y", "@notionhq/notion-mcp-server"],
"env": {
"OPENAPI_MCP_HEADERS": "{\"Authorization\": \"Bearer ntn_****\", \"Notion-Version\": \"2022-06-28\" }"
}
}
}
}
Zed
Add the following to your settings.json
{
"context_servers": {
"some-context-server": {
"command": {
"path": "npx",
"args": ["-y", "@notionhq/notion-mcp-server"],
"env": {
"OPENAPI_MCP_HEADERS": "{\"Authorization\": \"Bearer ntn_****\", \"Notion-Version\": \"2022-06-28\" }"
}
},
"settings": {}
}
}
}
Using Docker:
There are two options for running the MCP server with Docker:
Option 1: Using the official Docker Hub image:
Add the following to your .cursor/mcp.json
or claude_desktop_config.json
:
Using NOTION_TOKEN (recommended):
{
"mcpServers": {
"notionApi": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"-e", "NOTION_TOKEN",
"mcp/notion"
],
"env": {
"NOTION_TOKEN": "ntn_****"
}
}
}
}
Using OPENAPI_MCP_HEADERS (for advanced use cases):
{
"mcpServers": {
"notionApi": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"-e", "OPENAPI_MCP_HEADERS",
"mcp/notion"
],
"env": {
"OPENAPI_MCP_HEADERS": "{\"Authorization\":\"Bearer ntn_****\",\"Notion-Version\":\"2022-06-28\"}"
}
}
}
}
This approach:
Option 2: Building the Docker image locally:
You can also build and run the Docker image locally. First, build the Docker image:
docker compose build
Then, add the following to your .cursor/mcp.json
or claude_desktop_config.json
:
Using NOTION_TOKEN (recommended):
{
"mcpServers": {
"notionApi": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"-e",
"NOTION_TOKEN=ntn_****",
"notion-mcp-server"
]
}
}
}
Using OPENAPI_MCP_HEADERS (for advanced use cases):
{
"mcpServers": {
"notionApi": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"-e",
"OPENAPI_MCP_HEADERS={\"Authorization\": \"Bearer ntn_****\", \"Notion-Version\": \"2022-06-28\"}",
"notion-mcp-server"
]
}
}
}
Don't forget to replace ntn_****
with your integration secret. Find it from your integration configuration tab:
To install Notion API Server for Claude Desktop automatically via Smithery:
npx -y @smithery/cli install @makenotion/notion-mcp-server --client claude
The Notion MCP Server supports two transport modes:
The default transport mode uses standard input/output for communication. This is the standard MCP transport used by most clients like Claude Desktop.
# Run with default stdio transport
npx @notionhq/notion-mcp-server
# Or explicitly specify stdio
npx @notionhq/notion-mcp-server --transport stdio
For web-based applications or clients that prefer HTTP communication, you can use the Streamable HTTP transport:
# Run with Streamable HTTP transport on port 3000 (default)
npx @notionhq/notion-mcp-server --transport http
# Run on a custom port
npx @notionhq/notion-mcp-server --transport http --port 8080
# Run with a custom authentication token
npx @notionhq/notion-mcp-server --transport http --auth-token "your-secret-token"
When using Streamable HTTP transport, the server will be available at http://0.0.0.0:<port>/mcp
.
Authentication
The Streamable HTTP transport requires bearer token authentication for security. You have three options:
Option 1: Auto-generated token (recommended for development)
npx @notionhq/notion-mcp-server --transport http
The server will generate a secure random token and display it in the console:
Generated auth token: a1b2c3d4e5f6789abcdef0123456789abcdef0123456789abcdef0123456789ab
Use this token in the Authorization header: Bearer a1b2c3d4e5f6789abcdef0123456789abcdef0123456789abcdef0123456789ab
Option 2: Custom token via command line (recommended for production)
npx @notionhq/notion-mcp-server --transport http --auth-token "your-secret-token"
Option 3: Custom token via environment variable (recommended for production)
AUTH_TOKEN="your-secret-token" npx @notionhq/notion-mcp-server --transport http
The command line argument --auth-token
takes precedence over the AUTH_TOKEN
environment variable if both are provided.
Making HTTP Requests
All requests to the Streamable HTTP transport must include the bearer token in the Authorization header:
# Example request
curl -H "Authorization: Bearer your-token-here" \
-H "Content-Type: application/json" \
-H "mcp-session-id: your-session-id" \
-d '{"jsonrpc": "2.0", "method": "initialize", "params": {}, "id": 1}' \
http://localhost:3000/mcp
Note: Make sure to set either the NOTION_TOKEN
environment variable (recommended) or the OPENAPI_MCP_HEADERS
environment variable with your Notion integration token when using either transport mode.
Comment "Hello MCP" on page "Getting started"
AI will correctly plan two API calls, v1/search
and v1/comments
, to achieve the task
Add a page titled "Notion MCP" to page "Development"
Get the content of page 1a6b35e6e67f802fa7e1d27686f017f2
Build
npm run build
Execute
npx -y --prefix /path/to/local/notion-mcp-server @notionhq/notion-mcp-server
Publish
npm publish --access public
Please log in to share your review and rating for this MCP.
{ "mcpServers": { "notionApi": { "command": "npx", "args": [ "-y", "@notionhq/notion-mcp-server" ], "env": { "NOTION_TOKEN": "ntn_****" } } } }
Discover more MCP servers with similar functionality and use cases
by elie222
An AI‑powered email assistant that automates inbox management, enabling users to reach inbox zero fast by handling replies, labeling, archiving, unsubscribing, and providing analytics through a plain‑text prompt configuration.
by sooperset
MCP Atlassian is a Model Context Protocol (MCP) server that integrates AI assistants with Atlassian products like Confluence and Jira. It enables AI to automate tasks, search for information, and manage content within Atlassian ecosystems.
by ggozad
Interact with Ollama models through an intuitive terminal UI, supporting persistent chats, system prompts, model parameters, and MCP tools integration.
by nbonamy
A desktop AI assistant that bridges dozens of LLM, image, video, speech, and search providers, offering chat, generative media, RAG, shortcuts, and extensible plugins directly from the OS.
by GongRzhe
Provides tools for creating, editing, and enhancing PowerPoint presentations through a comprehensive set of MCP operations powered by python-pptx.
by GongRzhe
Creates, reads, and manipulates Microsoft Word documents through a standardized interface for AI assistants, enabling rich editing, formatting, and analysis capabilities.
by GongRzhe
Gmail-MCP-Server is a Model Context Protocol (MCP) server that integrates Gmail functionalities into AI assistants like Claude Desktop. It enables natural language interaction for email management, supporting features like sending, reading, and organizing emails.
by nspady
google-calendar-mcp is a Model Context Protocol (MCP) server that integrates Google Calendar with AI assistants. It enables AI assistants to manage Google Calendar events, including creating, updating, deleting, and searching for events.
by runebookai
Provides a desktop interface to chat with local or remote LLMs, schedule tasks, and integrate Model Context Protocol servers without coding.