Gemini-Data-Analysis-Research

This MCP server provides three main capabilities: data analysis with AI-assisted insights, generation and delivery of research-oriented content via email, and a thinking-generation feature for producing structured analyses and documents. The analyze-data tool accepts an input file (Excel or CSV), detects data types, runs statistical analyses, generates interactive visualizations, and produces HTML reports with optional large-file support. The generate-thinking tool creates research papers, technical documentation, data analyses, or BI reports based on a user prompt, and can optionally send the generated content via email. The send-email tool leverages Gemini-powered subject generation, supports multiple content types, attaches images, and delivers messages through SMTP with tracking and templates. To use these tools, ensure your environment is configured with your Gemini API key, a valid SMTP account, and the appropriate environment variables; then invoke the MCP endpoints via the JSON payloads described in the Quick Start sections of the documentation.

Usage examples (conceptual):

• Analyze data: Provide a base64-encoded file, file name, and desired output directory. The tool returns analysis results and a report with plots.
• Generate thinking: Provide a prompt and optional output directory to obtain a structured analysis or content file, then optionally send via email.
• Send email: Provide recipient, a subject prompt, plain text, optional HTML, and optional inline images to be attached and embedded in the email.

In Claude Desktop or your MCP client, you’ll typically pass a JSON payload like the examples in the README to routes named analyze-data, generate-thinking, or send-email. The server handles the orchestration with Gemini AI and your SMTP setup to deliver results and documents to stakeholders.

Prerequisites

• Node.js v16 or higher
• TypeScript tooling (for development/build)
• Claude Desktop (optional, for deployment with Claude integration)
• Google Gemini API Key
• SMTP Email Account (for email functionality)

Installation steps

1. Clone the repository and install dependencies:

git clone [your-repo-url]
cd gemini-data-analysis-email-generator
npm install

2. Create a .env file (or set environment variables) with required keys:

GEMINI_API_KEY=your_api_key_here
NODEMAILER_EMAIL=your.email@gmail.com
NODEMAILER_PASSWORD=your_app_password_here

3. Build the project (if using TypeScript source):

npm run build

4. Configure Claude Desktop (optional):

• Edit %AppData%/Claude/claude_desktop_config.json and add:

{
  "mcpServers": {
    "Gemini Data Analysis": {
      "command": "node",
      "args": ["path/to/gemini-data-analysis-email-generator/dist/index.js"],
      "cwd": "path/to/gemini-data-analysis-email-generator",
      "env": {
        "GEMINI_API_KEY": "your_api_key_here",
        "NODEMAILER_EMAIL": "your.email@gmail.com",
        "NODEMAILER_PASSWORD": "your_app_password_here"
      }
    }
  }
}

5. Run the server:

npm run start

6. (Optional) Development mode:

npm run dev

Tips and notes:

• Store API keys securely and avoid committing .env to version control.
• For Gmail, use an app password rather than your main password; ensure less secure app access is managed appropriately if not using app passwords.
• Ensure your SMTP settings and firewall allow outbound connections to the SMTP server.
• When handling large datasets, adjust outputDir and consider chunking inputs to stay within memory and time limits.
• The environment variables GEMINI_API_KEY, NODEMAILER_EMAIL, and NODEMAILER_PASSWORD are required for full functionality; verify they are loaded before starting the MCP server.
• Enable DEBUG=true in .env for verbose logs during troubleshooting.
• If you encounter API key errors, re-check the .env file and ensure it’s loaded by your process at startup.