Politwit1984 Second Opinion MCP Server

🚀 Second Opinion MCP Server

A coding problem-solving server that combines multi-source AI assistance, integrating the following features:

• Insights from Google Gemini AI
• Answers from the Stack Overflow community
• Analytical capabilities of Perplexity AI

🚀 Quick Start

The Second Opinion MCP Server is a powerful tool that integrates multiple AI sources to solve coding problems. It combines insights from Google Gemini AI, answers from the Stack Overflow community, and the analytical capabilities of Perplexity AI.

✨ Features

• Provide detailed programming problem solutions by integrating context information from multiple sources.
• Automatically recognize the programming language based on the file extension.
• Intelligently extract and format code snippets.
• Automatically generate structured Markdown solution reports.
• Analyze the Git dependencies and context of project files.

📦 Installation

1. Install Dependencies: Run the npm install command to install all required packages.
2. Build the Server: Use npm run build to build the MCP server.
3. Configure Environment Variables:
   • Configuration file path: src/config.ts

   export const API_KEYS = {
     GEMINI_API_KEY: 'your_api_key_here',
     PERPLEXITY_API_KEY: 'your_api_key_here',
     STACK_OVERFLOW_API_KEY: 'your_api_key_or_empty' // Optional, anonymous access by default
   };
   

💻 Usage Examples

Basic Usage

Solve problems by calling the get_second_opinion tool with the following input format:

{
  "goal": "Goal description",
  "error": "Error message",
  "code": "Related code",
  "solutionsTried": "Array of tried solutions",
  "filePath": "File path"
}

Advanced Usage

Example Scenario: Solve the cleanup problem of useEffect in React

{
  "goal": "Fix the error caused by the cleanup method of useEffect in React",
  "error": "TypeError: Cannot read property 'state' of undefined",
  "code": "const [count, setCount] = useState(0);\nuseEffect(() => {\n  const interval = setInterval(() => {\n    setCount(prev => prev + 1);\n  }, 1000);\n  return () => clearInterval(interval);\n}, []);",
  "solutionsTried": ["Remove the cleanup function", "Check for undefined in the effect"],
  "filePath": "src/components/Timer.tsx"
}

Solution Examples

Basic Problem Analysis

The error message indicates that setCount may not be correctly bound during the cleanup phase. Using useRef and useCallback can effectively solve such problems.

Provided Solutions

Solution 1: Use useRef to Save the Cleanup Function

import React, { useEffect, useState, useRef } from 'react';

export default function Timer() {
  const [count, setCount] = useState(0);
  const cleanupRef = useRef<NodeJS.Timeout>();

  useEffect(() => {
    cleanupRef.current = setInterval(() => {
      setCount(prev => prev + 1);
    }, 1000);

    return () => {
      if (cleanupRef.current) {
        clearInterval(cleanupRef.current);
        cleanupRef.current = undefined;
      }
    };
  }, []);

  return <div>Current count: {count}</div>;
}

Solution 2: Use useCallback to Handle State

import React, { useEffect, useState, useCallback } from 'react';

export default function Timer() {
  const [count, setCount] = useState(0);
  const callback = useCallback((prev: number) => prev + 1, []);

  useEffect(() => {
    const interval = setInterval(callback, 1000);
    return () => clearInterval(interval);
  }, [callback]);

  return <div>Current count: {count}</div>;
}

📚 Documentation

Best Practices

• Dependency Management: Use useRef to maintain state and avoid unnecessary re-rendering.
• Performance Optimization: Use useMemo and useCallback appropriately to cache calculation results.
• Error Handling: Add sufficient error catching and handling logic in WebSocket or asynchronous operations.
• Testing Considerations: Write unit tests to cover major functions and mock third-party API calls.

Project Structure

src/
├── config.ts        # Configuration file containing API keys for each service
├── fileUtils.ts     # File operation functions and language detection logic
├── index.ts         # Application entry point
├── perplexity.ts   # Integration code for Perplexity AI
├── server.ts       # Core implementation of the MCP server
├── stackOverflow.ts # Call handling for the Stack Overflow API
└── types.ts        # Define TypeScript interfaces and types used in the project

Known Issues

Refer to the errors.md file for a detailed list of issues.