perplexity-search

Retrieve real-time web search results with source citations via OpenRouter.

Updated Feb 13, 2026

Search Tags:#citations #openrouter #perplexity #web-search #model-selection #grounded-answers #liteLLM

Ready to run

npx skills add https://github.com/mwathiben/PropManager --skill perplexity-search-mwathiben

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: perplexity-search
Source: https://github.com/mwathiben/PropManager/tree/main/.claude/skills/perplexity-search
Command: npx skills add https://github.com/mwathiben/PropManager --skill perplexity-search-mwathiben

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Real-time AI-powered web search results with source citations, enabling users to quickly obtain current information beyond model knowledge.

Core Features & Use Cases

Real-time web search across Perplexity models via OpenRouter.
Single API key access to multiple models and cost-guided model selection.
Citations and verifiable sources for grounded answers, ideal for literature reviews and technical research.

Quick Start

Run a query with the perplexity_search.py script to retrieve real-time web results.

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 an OpenRouter API key?▼

Real-time web search with source citations is achieved by routing queries to Perplexity models via OpenRouter. You configure your OPENROUTER_API_KEY, run the perplexity_search.py script, and retrieve grounded answers with verifiable sources.

What is the best way to conduct technical research using Perplexity models through LiteLLM?▼

Conducting technical research with Perplexity models through LiteLLM involves executing the perplexity_search.py CLI workflow. This routes your query to openrouter/perplexity/* endpoints, providing current, web-grounded information with citations for literature reviews.

Can I configure max_tokens and temperature when running web-grounded queries across multiple Perplexity models?▼

You can configure max_tokens and temperature when running web-grounded queries across multiple Perplexity models. The CLI workflow in perplexity_search.py allows you to set these parameters while using a single OpenRouter API key for model routing.

Do I need a separate API key for each Perplexity model to access grounded web search?▼

You do not need a separate API key for each Perplexity model. A single OPENROUTER_API_KEY provides access to multiple models, enabling cost-guided model selection and routing to openrouter/perplexity/* for grounded web search results.

How does cost-guided model selection work for real-time web searches with OpenRouter?▼

Cost-guided model selection for real-time web searches allows users to choose among multiple Perplexity models available via OpenRouter. By routing queries through openrouter/perplexity/*, you can balance cost and performance while retrieving grounded answers.

Why use Perplexity models via OpenRouter instead of standard LLM APIs for scientific research?▼

Using Perplexity models via OpenRouter provides real-time web search results with source citations, unlike standard LLM APIs limited to training data. This enables verifiable, grounded answers essential for scientific and technical research.