perplexity-search

Search the web via Perplexity models through LiteLLM and OpenRouter with source citations.

8|Updated Jan 13, 2026

Search Tags:#research #web search #citations #real-time data #literature search #perplexity

Ready to run

npx skills add https://github.com/hxk622/TokenDance --skill perplexity-search-hxk622

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: perplexity-search
Source: https://github.com/hxk622/TokenDance/tree/main/backend/app/skills/builtin/scientific/research-tools/perplexity-search
Command: npx skills add https://github.com/hxk622/TokenDance --skill perplexity-search-hxk622

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 provides access to up-to-date information from the web, overcoming the knowledge cut-off limitations of standard AI models and delivering grounded answers with source citations.

Core Features & Use Cases

Real-time Web Search: Conduct AI-powered searches for current events, recent scientific literature, and any information not present in model training data.
Source Citations: Get answers grounded in web sources with verifiable citations.
Multiple Perplexity Models: Access various Perplexity models (Sonar Pro, Sonar Pro Search, Sonar Reasoning Pro) via a single OpenRouter API key.
Use Case: When researching the latest advancements in quantum computing or seeking up-to-the-minute clinical trial results, this skill provides accurate, cited information.

Quick Start

Use the perplexity-search skill to find 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 web search results with source citations using AI?▼

To get real-time web search results with source citations, you can use an AI-powered search skill that queries the web via Perplexity models through LiteLLM and OpenRouter. This approach retrieves current information grounded with verifiable sources.

Can I access recent scientific literature beyond a model's knowledge cutoff?▼

Yes, you can access recent scientific literature beyond a model's knowledge cutoff by performing AI-powered web searches. This retrieves up-to-the-minute data like clinical trial results and recent advancements with source citations.

Do I need an OpenRouter API key to use Perplexity models for web search?▼

Yes, you need an OpenRouter API key to access multiple Perplexity models such as Sonar Pro and Sonar Reasoning Pro. This single key manages the API routing required to execute the web search.

What is the best way to find current information for fact-verification tasks?▼

The best way to find current information for fact-verification is using an AI web search skill that retrieves real-time data with citations. This grounds your answers in verifiable web sources, ensuring accuracy.

Does this AI web search approach work with LiteLLM for research analysis?▼

Yes, this AI web search approach works with LiteLLM to support research and analysis tasks. It uses LiteLLM to connect with Perplexity models, delivering real-time, web-grounded information with source citations.

What are the limitations of using AI models for real-time web search?▼

The main limitation of standard AI models for real-time web search is their knowledge cutoff. This skill overcomes that limitation by querying the web via Perplexity models, though it requires an active OpenRouter API key.