Perplexity MCP Server Gfh

🚀 Project Overview

This is a project built on the Model Context Protocol, primarily designed to integrate with Perplexity. Through this project, we can leverage Perplexity's powerful search engine and large language model (LLM) to provide users with more intelligent and efficient search-enhanced Q&A capabilities.

✨ Features

• Search-Enhanced Q&A: Users can obtain answers based on Perplexity's search results through natural language queries.
• Multi-Parameter Support: Allows customizing search behaviors, such as specifying related questions to return, filtering search time ranges, and domains.
• Scalability: The framework is designed flexibly, facilitating the addition of new tools or functional modules.

🔧 Technical Details

The overall architecture of the project follows a clear hierarchical principle:

                    ┌──────────────┐
                    │    Client     │
                    └──────┬───────┘
                           │
                    ┌──────▼───────┐
                    │   Handler    │
                    └──────┬───────┘
                           │
                    ┌──────▼───────┐
                    │  Server      │
                    └──────┬───────┘
                           │
                    ┌──────▼───────┐
                    │ Services     │
                    └──────┬───────┘
                           │
                    ┌──────▼───────┐
                    │   Config     │
                    └──────┬───────┘
                           │
                    ┌──────▼───────┐
                    │  Database    │
                    └──────────────┘

🚀 Quick Start

Installation

1. Install dependencies:

npm install mcp-ts-template @perplexity/api

2. Initialize configuration: Create a config.ts file and add the following content:

export const config = {
  api_key: 'your-perplexity-api-key',
  search_context_size: 5,
  // Other custom configurations
};

3. Start the service: Run the command:

npm run dev

💻 Usage Examples

perplexity_search Tool

This tool performs search-enhanced queries through the Perplexity API and is mainly used to answer questions based on web search results.

Input Parameters

Example Usage

const result = await perplexity_search({
  query: "What is artificial intelligence?",
  return_related_questions: true,
  search_recency_filter: 'week',
  search_domain_filter: ['example.com']
});

Development Guide

Project Structure

• src/: Contains core business logic and modules.
• config/: Stores the application's configuration files.
• types/: Defines custom types and interfaces.

Quick Development Process

1. Create a new functional module:

   • Create a new component file in the src/components directory.
   • Implement the relevant business logic and ensure it complies with project specifications.

2. Test new features:

   • Write and run test cases using a unit testing framework (e.g., Jest).
   • Ensure that all new features pass the test verification.

3. Deployment and Release:

   • Run npm run build locally to generate build files.
   • Commit the code to the version control system and release it according to semantic versioning.

Project Maintenance

• Logging: Add appropriate log outputs at key business logic points for debugging and monitoring.
• Error Handling: Provide a comprehensive error capture and recovery mechanism for all API calls and external service interactions.
• Performance Optimization: Regularly review the code and optimize potential performance bottlenecks.

📄 License

This project is licensed under the MIT License, which allows free use, modification, and distribution, provided that the copyright notice is retained.