perplexity-mcp MCP server

@jsonallen

About perplexity-mcp MCP server

A Model Context Protocol (MCP) server that provides web search functionality using Perplexity AI's API.

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "perplexity-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@smithery/cli",
        "install",
        "perplexity-mcp",
        "--client",
        "claude"
      ]
    }
  }
}

Tools

Search the web using Perplexity AI

Overview

What is perplexity-mcp?

perplexity-mcp is a Model Context Protocol (MCP) server that provides web search functionality using the Perplexity AI API. It is designed to work with the Anthropic Claude desktop client.

How to use perplexity-mcp?

Install via Smithery with npx -y @smithery/cli install perplexity-mcp --client claude or manually using uvx with the required PERPLEXITY_API_KEY environment variable and optional PERPLEXITY_MODEL. Add the tool to your Cursor or Claude Desktop configuration file to invoke search prompts and tools.

Key features of perplexity-mcp

Single prompt perplexity_search_web with required query argument.
Optional recency parameter to filter results by day, week, month, or year.
Single tool perplexity_search_web returning search results from Perplexity.
Multiple model choices configurable via PERPLEXITY_MODEL environment variable.
Works with Cursor and Claude Desktop clients via MCP protocol.

Use cases of perplexity-mcp

Search the web for recent news or topics using natural language.
Filter results by recency (last day, week, month, or year).
Integrate web search capabilities into Claude-powered workflows.
Automate research queries by calling the tool programmatically.

FAQ from perplexity-mcp

What runtime does perplexity-mcp require?

It requires UV, a fast Python package and project manager. Install via brew install uv on macOS or the installer script for Linux/Windows.

How do I get a Perplexity API key?

Obtain an API key from Perplexity and set it as the PERPLEXITY_API_KEY environment variable.

What model does perplexity-mcp use by default?

The default model is sonar. You can override it by setting the PERPLEXITY_MODEL environment variable to any supported model like sonar-pro or sonar-deep-research.

How do I verify the server is working?

Use a prompt like “search the web for news about OpenAI in the past week” in Claude Desktop. An alert box will appear to confirm tool usage—click “Allow for this chat”.

Where is the configuration file for Claude Desktop?

On macOS it is at ~/Library/Application Support/Claude/claude_desktop_config.json. On Windows it is at %APPDATA%/Claude/claude_desktop_config.json.

Frequently asked questions

What runtime does perplexity-mcp require?

It requires UV, a fast Python package and project manager. Install via `brew install uv` on macOS or the installer script for Linux/Windows.

How do I get a Perplexity API key?

Obtain an API key from [Perplexity](https://perplexity.ai) and set it as the `PERPLEXITY_API_KEY` environment variable.

What model does perplexity-mcp use by default?

The default model is `sonar`. You can override it by setting the `PERPLEXITY_MODEL` environment variable to any supported model like `sonar-pro` or `sonar-deep-research`.

How do I verify the server is working?

Use a prompt like “search the web for news about OpenAI in the past week” in Claude Desktop. An alert box will appear to confirm tool usage—click “Allow for this chat”.

Where is the configuration file for Claude Desktop?

On macOS it is at `~/Library/Application Support/Claude/claude_desktop_config.json`. On Windows it is at `%APPDATA%/Claude/claude_desktop_config.json`.

Comments

Basic information

More Search MCP servers

Tavily Mcp

tavily-ai

Production ready MCP server with real-time search, extract, map & crawl.

2.2K

Google Search Console MCP Server

ahonn

A Model Context Protocol (MCP) server providing access to Google Search Console

239

mcp-omnisearch

spences10

🔍 A Model Context Protocol (MCP) server providing unified access to multiple search engines (Tavily, Brave, Kagi, Exa), AI tools (Kagi FastGPT, Exa, Linkup), and content extraction services (Firecrawl, Tavily, Kagi). Includes GitHub search. All through a single interface.

332