perplexity-mcp  by jsonallen

Web search integration for AI assistants

Created 1 year ago
290 stars

Top 90.8% on SourcePulse

GitHubView on GitHub
Project Summary

This project offers a Model Context Protocol (MCP) server that bridges Perplexity AI's web search capabilities with desktop AI clients, notably Anthropic Claude and Cursor. It empowers users to seamlessly integrate real-time web information into their AI interactions, enhancing the AI's ability to answer questions requiring up-to-date external data.

How It Works

The server implements a perplexity_search_web tool, allowing AI clients to query Perplexity AI's API for web searches. Users can specify a search query and an optional recency parameter to filter results by time period ('day', 'week', 'month', 'year'). The system supports various Perplexity models, such as 'sonar' and 'sonar-pro', enabling users to select models optimized for different research needs and context lengths. This architecture directly injects web search functionality into the AI's operational context.

Quick Start & Requirements

  • Installation: Primarily via Smithery: npx -y @smithery/cli install perplexity-mcp --client claude.
  • Prerequisites: Requires the UV package manager. Installation instructions are provided for macOS (Homebrew: brew install uv), Linux/macOS (curl script), and Windows (PowerShell).
  • Configuration: A Perplexity AI API key is mandatory, set via the PERPLEXITY_API_KEY environment variable. An optional PERPLEXITY_MODEL variable allows choosing from available Perplexity models. The server must be registered in client configuration files (mcp.json for Cursor, claude_desktop_config.json for Claude Desktop).
  • Setup: Involves obtaining an API key and configuring the client's JSON file, typically a quick process.

Highlighted Details

  • Enables natural language web searches directly within AI chat interfaces.
  • Supports time-based filtering of search results ('day', 'week', 'month', 'year').
  • Offers flexibility in choosing Perplexity AI models for tailored search performance.
  • Designed for seamless integration with AI desktop clients like Cursor and Claude Desktop.

Maintenance & Community

The provided README snippet does not contain information regarding maintainers, community channels (e.g., Discord, Slack), or project roadmaps.

Licensing & Compatibility

The specific open-source license is not mentioned in the provided documentation. Compatibility is explicitly stated for Cursor and Claude Desktop clients.

Limitations & Caveats

The project's functionality is dependent on Perplexity AI's API availability and terms of service. Details on error handling, rate limiting, or advanced configuration are not elaborated. The absence of a specified license may pose a barrier for certain commercial or closed-source integration scenarios.

Health Check
Last Commit

1 year ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
0
Star History
6 stars in the last 30 days

Explore Similar Projects

Starred by Chip Huyen Chip Huyen(Author of "AI Engineering", "Designing Machine Learning Systems"), Zack Li Zack Li(Cofounder of Nexa AI), and
12 more.

search_with_lepton by leptonai

0.0%
8k
Conversational search engine demo
Created 2 years ago
Updated 4 months ago
Feedback? Help us improve.