Perplexity AI Search

Integrates with Perplexity AI's search capabilities to enable AI-powered web search queries using TypeScript and the OpenAI SDK.

Installation Instructions

README: https://github.com/mkusaka/mcp-server-perplexity

Perplexity AI MCP Server

An MCP server implementation that integrates Perplexity AI's API, providing advanced search capabilities with multiple model options.

Features

• Search using Perplexity AI's models
• Support for all official Sonar models
• Configurable result count
• Detailed error handling and logging
• MCP Inspector compatible

Available Models

• sonar-reasoning-pro (127k context)
• sonar-reasoning (127k context)
• sonar-pro (200k context)
• sonar (127k context)

Installation

pnpm install
pnpm build

Configuration

API Key

1. Sign up for a Perplexity AI account
2. Get your API key from the dashboard
3. Set the environment variable:

export PERPLEXITY_API_KEY=your_api_key_here

Usage

Direct Execution

node dist/index.js
# or if you made it executable
./dist/index.js

Development

pnpm dev

Testing with MCP Inspector

pnpm inspect

Tool Reference

perplexity_search

Performs a search using Perplexity AI's models.

Parameters:

• query (string, required): The search query
• model (string, optional): Model to use (default: "sonar")
  • Available options: sonar-reasoning-pro, sonar-reasoning, sonar-pro, sonar
• count (number, optional): Number of results (1-10, default: 5)

Development

Project Structure

src/perplexity/
├── index.ts        # Main server implementation
├── lib/
│   └── logger.ts   # Logging configuration
└── adr.md         # Architectural decisions

Logging

Logs are written to perplexity-mcp.log in the project root directory.

License

MIT

Contributing

1. Fork the repository
2. Create your feature branch
3. Commit your changes
4. Push to the branch
5. Create a new Pull Request