Gemini Data Analysis & Research

🚀 Libra Data Gemini Data Analysis and Research MCP Server

The Libra Data Gemini Data Analysis and Research MCP Server is an integrated platform designed for processing and analyzing Gemini data. It is developed based on Node.js and TypeScript, supporting various data analysis tasks and achieving efficient data processing through flexible interfaces.

🚀 Quick Start

Installation and Configuration

1. Clone the code repository to your local machine.
2. Create a .env file and set the following environment variables:
   • GEMINI_API_KEY: Libra Data API key
   • NODEMAILER_EMAIL: Email address of the sender
   • NODEMAILER_PASSWORD: Application password for the email

Start the Service

Run the following command to start the MCP server:

npm run start

✨ Features

1. Data Analysis Tools

• File Parsing and Data Extraction: Supports .xlsx, .xls, and .csv file formats.
• Basic Statistical Analysis: Provides basic data summarization and statistical functions.
• Advanced Analysis Functions: Includes data visualization and in - depth data analysis.

2. Email Sending Tools

• Email Content Generation: Automatically generates email subjects and bodies using the Gemini AI model.
• Multi - version Support: Supports plain text and HTML - formatted email content.
• Attachment Management: Allows adding and managing image attachments.

3. Thinking Generation Tools

• Analysis Prompt Processing: Parses users' analysis requests and generates corresponding outputs.
• Result Storage: Saves the thinking process and results to the specified directory.

💻 Usage Examples

Data Analysis Tool Interface

interface AnalysisDataParams {
  fileData: string;         // Base64 - encoded file content
  fileName: string;         // File name (must be in .xlsx, .xls, or .csv format)
  analysisType: 'basic' | 'detailed';  // Analysis type
  outputDir?: string;      // Optional output directory
}

Email Sending Tool Interface

interface SendEmailParams {
  to: string;              // Recipient's email address
  subjectPrompt: string;   // Prompt the Gemini model to generate the email subject
  text: string;            // Plain - text version of the email content
  html?: string;           // HTML - formatted email content (optional)
  images?: {               // Optional image attachments
    name: string;          // Image file name
    data: string;          // Base64 - encoded image data
  }[];
}

Thinking Generation Tool Interface

interface GenerateThinkingParams {
  prompt: string;           // Analysis prompt
  outputDir?: string;       // Optional output directory
}

📚 Documentation

Output Structure

• data/: Stores data parsing and statistical results.
• thinking/: Saves the thinking process records of Gemini.
• emails/: Records email sending logs.

Script Commands

• npm run build: Compiles TypeScript code to JavaScript.
• npm run start: Starts the MCP server.
• npm run dev: Runs in development mode with hot - reloading support.

Environment Variable Configuration

• GEMINI_API_KEY: Libra Data API key
• NODEMAILER_EMAIL: Email address used for sending emails
• NODEMAILER_PASSWORD: Application password for the email (when using Gmail, use the app - specific password)

🔧 Technical Details

Security Precautions

1. Key Management:

   • Avoid storing API keys in plain text in the code.
   • Use environment variables or encrypted configuration files to manage keys.

2. Permission Control:

   • Ensure that server ports and resource access permissions are properly set.
   • Regularly review and update access policies.

3. Log Monitoring:

   • Enable detailed logging functionality.
   • Regularly audit log files to identify abnormal behavior.

📄 License

This software follows the MIT open - source license. For detailed license terms, please refer to the LICENSE file in the project repository.

Project Author

• Developer: To be filled
• Contact Information: To be filled