Perplexity Insight

🚀 Perplexity-Insight-MCP Server

A server implementation of MCP (Model Context Protocol) for interacting with the Perplexity AI API, enabling advanced question-answering capabilities through the standardized Model Context Protocol.

🚀 Quick Start

This is a server implementation of MCP for interacting with the Perplexity AI API. It offers advanced question-answering capabilities through the standardized Model Context Protocol.

✨ Features

• Seamless integration with the Perplexity AI API.
• Support for different Perplexity models (sonar-reasoning, sonar-pro, sonar-deep-research).
• Customizable system prompts and user queries.
• Proper error handling and response formatting.
• Protection against rate limits.
• Easy integration with the Windsurf IDE.

📦 Installation

Requirements

• Node.js 18+
• Perplexity API key

Steps

npm install

Environment Variables

Create a .env file containing the following variable:

PERPLEXITY_API_KEY=your_api_key_here

💻 Usage Examples

Basic Usage

Run the server:

npm start

API Tools

The server exposes the following tools:

1. perplexity_ask - Send questions directly to Perplexity AI.
2. perplexity_search - Execute search queries using Perplexity AI.

Changing the Model

Both tools support the following Perplexity models:

• sonar-reasoning (default) - Perplexity's reasoning-focused model, best for general questions.
• sonar-pro - An enhanced model with improved capabilities for professional use cases.
• sonar-deep-research - Specialized for in-depth research and complex queries.

To specify a model when using the tools, include the model parameter in the request:

Ask Perplexity using sonar-deep-research: What are the latest advancements in quantum computing?

You can also customize the system prompt and the maximum number of tokens:

Search through Perplexity using sonar-pro with a system prompt of "You are a helpful research assistant" and a maximum token count of 2000: Latest developments in renewable energy

Tool Response Format

The server follows the tool response format of the MCP specification:

{
  content: [
    {
      type: "text",
      text: "Answer content from Perplexity AI"
    }
  ],
  isError: false // or true if an error occurs
}

Windsurf Integration

Setup in Windsurf

1. Build the server:

   npm run build
   

2. Open Windsurf and navigate to the settings.
3. Find the "AI Settings" or "Model Context Protocol" section.
4. Add a new MCP server with the following details:
   • Name: Perplexity Insight
   • Type: Local process
   • Command: Path to your Node.js executable
   • Arguments: Path to your compiled index.js file
   • Working Directory: Path to your project directory
   • Environment Variables: Ensure it includes PERPLEXITY_API_KEY=your_api_key_here
5. Enable the server and restart Windsurf if necessary.

Example Configuration

Here is an example configuration for the mcp_config.json file:

"perplexity-ask": {
  "command": Path to Node.js executable,
  "args": [
    "node",
    "index.js"
  ],
  "env": {
    "PERPLEXITY_API_KEY": "your_api_key_here"
  }
}

Troubleshooting

• Problem: Unable to connect to the Perplexity API. Solution: Check if the API key is correct and ensure a stable internet connection.

Development

npm run dev

📄 License

This project is licensed under the MIT License.