Perplexity MCP Server

Example Use Cases

• 1

  Send direct natural-language queries to the Perplexity API and receive structured responses via MCP tools.

• 2

  Perform web searches and retrieve search results using Perplexity's search capabilities within an MCP-compatible environment.

• 3

  Run locally (STDIO) or remotely (SSE) and integrate Perplexity-powered tools into clients like Claude Desktop or Cursor.

Description

This project provides a Model Context Protocol (MCP) server that integrates with the Perplexity API using the OpenAI SDK and a custom base URL. It exposes two primary tools — perplexity-query for direct queries and perplexity-search for web searches — and supports local STDIO and remote SSE transports for agent clients. The server includes detailed logging, .env-based configuration for the Perplexity API key, and Docker deployment options (including a multi-arch image and deploy scripts) for easy hosting. It is aimed at enabling MCP-compatible agents to leverage Perplexity's search and query capabilities within their workflows.