perplexity-search

Perform real-time web searches with Perplexity models and source citations.

Updated Dec 17, 2025
One-click install
npx skills add https://github.com/robotlearning123/claude-scientific-skills --skill perplexity-search-robotlearning123
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: perplexity-search
Source: https://github.com/robotlearning123/claude-scientific-skills/tree/main/scientific-skills/perplexity-search
Command: npx skills add https://github.com/robotlearning123/claude-scientific-skills --skill perplexity-search-robotlearning123

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires litellm, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill solves the problem of finding current information, recent scientific literature, and grounded answers with source citations by leveraging AI-powered web searches using Perplexity models.

Core Features & Use Cases

  • Real-time Web Searches: Access current information and recent scientific literature beyond the model's knowledge cutoff.
  • Source Citations: Get answers grounded in web sources with source citations.
  • Multiple Models: Access multiple Perplexity models including Sonar Pro, Sonar Pro Search, and Sonar Reasoning Pro through a single API key.
  • Use Case: Conduct literature searches across multiple domains, perform domain-specific research, and compare current approaches or technologies.

Quick Start

Use the perplexity-search skill to search for "What are the latest developments in CRISPR gene editing?".

Frequently Asked Questions about perplexity-search

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I get real-time information with source citations for scientific literature?

AI-powered web searches retrieve real-time information and recent scientific literature, providing answers grounded in web sources with source citations. This allows you to access current data beyond a model's knowledge cutoff for fact verification.

Do I need an OpenRouter API key to perform AI-powered web searches?

Yes, you need an OpenRouter API key to perform AI-powered web searches using Perplexity models. The Skill requires LiteLLM for API access to route requests to multiple Perplexity models for varying complexity and cost.

Can I access multiple Perplexity models through a single API key for domain-specific research?

Yes, you can access multiple Perplexity models including Sonar Pro, Sonar Pro Search, and Sonar Reasoning Pro through a single OpenRouter API key. This supports domain-specific research, literature searches, and current technology comparisons.

What is the best way to find current developments beyond a model's knowledge cutoff?

The best way to find current developments is using AI-powered web searches focused on real-time information retrieval. This approach fetches current information and recent scientific literature, returning grounded answers with source citations.

How does LiteLLM work with Perplexity models for technical queries?

LiteLLM works with Perplexity models by providing standardized API access for technical queries. It routes requests through an OpenRouter API key to perform AI-powered web searches, retrieving real-time information and source citations.

Are there limitations when using AI-powered web searches for fact verification?

AI-powered web searches for fact verification depend on the availability of current web sources and the selected Perplexity model's complexity and cost. Results are constrained by real-time information retrieval limits and the scope of accessible scientific literature.