Research

Provides an MCP server that retrieves survey data from Notion and creates survey pages in Notion via Claude Desktop Client.

python

GitHub Stars

python

Language

4 months ago

First Indexed

3 weeks ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "h-yanagawa-research-mcp-server": {
      "command": "uv",
      "args": [
        "run",
        "--with",
        "mcp[cli]",
        "--with",
        "notion-client",
        "--with",
        "pydantic",
        "mcp",
        "run",
        "/path/to/research-mcp-server/server.py"
      ],
      "env": {
        "NOTION_TOKEN": "YOUR_NOTION_TOKEN",
        "NOTION_ROOT_DATABASE_ID": "YOUR_NOTION_DATABASE_ID"
      }
    }
  }
}

You can run a dedicated MCP server that connects Claude’s desktop client to a Notion-backed survey workflow. This server lets you retrieve survey data from Notion and create Notion survey pages, all orchestrated through a lightweight local process you start from your Claude Desktop Client.

How to use

Once the MCP server is running, you interact with it through your MCP client inside the Claude Desktop Client. Start by launching the Research MCP Server, then ask Claude to perform a survey task. Claude will communicate with your Notion-backed data source, fetch responses, and display or aggregate results in your client. Use the server to centralize survey workflows, store results in Notion, and generate pages that reflect survey outcomes for quick review.

How to install

Prerequisites you need before starting are clearly defined: you must have Python 3.12.1 or later, uv 0.5.25 or later, and Claude Desktop Client installed on your machine.

Step 1: Prepare your Notion access

Obtain a Notion token from Notion Integrations.
Create a database page in Notion and capture the database ID from the page URL (the last segment of the URL). You will reference these values in your environment file.

Step 2: Create environment configuration

Create a file named .env in your project directory with your Notion credentials and database ID:

NOTION_ROOT_DATABASE_ID=[YOUR_NOTION_DATABASE_ID]```
- This file provides the credentials the MCP server will use to access Notion.

Step 3: Add the MCP server definition to Claude Desktop client configuration
- You will insert the following snippet into your Claude Desktop configuration under the `mcpServers` section. This defines how to start the MCP server using the local `uv` runner and the server script.

{ ..., "mcpServers": { "Research MCP Server": { "command": "uv", "args": [ "run", "--with", "mcp[cli]", "--with", "notion-client", "--with", "pydantic", "mcp", "run", "/path/to/research-mcp-server/server.py" ] } } }


Step 4: Start the server via Claude Desktop Client
- Restart or refresh the Claude Desktop Client to launch the Research MCP Server. The server will start based on the command and arguments you configured above.

Step 5: Run a survey
- With the server running, ask Claude to perform a survey and review the results. The data is retrieved from Notion and presented by Claude for your review and analysis.

## Additional notes

The MCP server integration relies on a Notion token and a root database ID. Ensure these are kept secure and only accessible by your local environment. If you need to change the Notion database, update the values in the `.env` file and restart the server.

## Available tools

### retrieve\_survey

Fetch survey data from Notion and load it into the MCP workflow so Claude can process and summarize results.

### create\_survey\_pages

Create or update Notion pages to reflect survey results, enabling easy review and sharing of insights.

### start\_server

Launch and manage the Research MCP Server from the Claude Desktop Client, wiring together the MCP runtime with Notion integration.

Built by

VeilStrat

AI signals for GTM teams