MCP Summarization Functions

🚀 MCP Server Tool for Command Execution, Summarization, and Text Analysis

This project is an MCP server tool designed for executing commands, summarizing files and directories, and conducting text analysis. It supports multiple output formats (e.g., text, JSON, Markdown, and outline) and can generate detailed information on security analysis, API surface view, error handling, dependencies, or type definitions based on specified focus areas.

🚀 Quick Start

📦 Installation

1. Install via npm Add this package as a dependency to your project using the following command:

   npm install mcp-summarization-functions
   

2. Integrate into MCP Configuration File Add the server to your MCP configuration file:

   {
     "mcpServers": {
       "SUMMARY_TOOL": {
         "command": "node",
         "args": ["./node_modules/mcp-summarization-functions/build/index.js"],
         "env": {
           "PROVIDER": "ANTHROPIC", // or other AI providers
           "API_KEY": "your-api-key",
           "MODEL_ID": "claude-3-5-sonnet"
         }
       }
     }
   }
   

✨ Features

1. Summarize Command Output

{
  command: string,    // The command to be executed
  cwd: string,       // The working directory for command execution
  
  // Optional parameters
  hint?: string,      // Analysis focus: "Security Analysis" | "API Surface" | "Error Handling" | "Dependencies" | "Type Definitions"
  outputFormat?: string  // Output format: "Text" | "JSON" | "Markdown" | "Outline" (default: "Text")
}

2. Summarize File Content

{
  paths: string[],    // An array of file paths to be summarized (relative to the working directory)
  cwd: string,       // The working directory for resolving file paths
  
  // Optional parameters
  hint?: string,      // Analysis focus: "Security Analysis" | "API Surface" | "Error Handling" | "Dependencies" | "Type Definitions"
  outputFormat?: string  // Output format: "Text" | "JSON" | "Markdown" | "Outline" (default: "Text")
}

3. Get Directory Structure Overview

{
  path: string,       // The directory path to be summarized (relative to the working directory)
  cwd: string,       // The working directory for resolving the directory path
  
  // Optional parameters
  recursive?: boolean,  // Whether to include subdirectories, suitable for handling deep directory structures
  hint?: string,       // Analysis focus: "Security Analysis" | "API Surface" | "Error Handling" | "Dependencies" | "Type Definitions"
  outputFormat?: string   // Output format: "Text" | "JSON" | "Markdown" | "Outline" (default: "Text")
}

4. Summarize Arbitrary Text Content

{
  content: string,    // The plain text content to be summarized
  type: string,       // The content type (e.g., "Log Output", "API Response")
  
  // Optional parameters
  hint?: string,      // Analysis focus: "Security Analysis" | "API Surface" | "Error Handling" | "Dependencies" | "Type Definitions"
  outputFormat?: string  // Output format: "Text" | "JSON" | "Markdown" | "Outline" (default: "Text")
}

5. Get Full Content

{
  id: string         // The ID of the stored content
}

🔧 Technical Details

The following environment variables can be used to customize the server behavior:

💻 Usage Examples

Basic Usage

• Summarize Command Output

await mcpClient.executeCommand("npm install", { cwd: "./my-project" }, {
  hint: "Security Analysis",
  outputFormat: "JSON"
});

• Get File Summary

const result = await mcpClient.getFileSummary(["./src/**/*"], { cwd: "." }, {
  hint: "Dependencies", 
  outputFormat: "Markdown"
});

📄 License

This project is licensed under the MIT License. For more details, please refer to the LICENSE file.