perplexity

Deliver real-time web search and chat completions via Sonar APIs.

12|4|Updated Feb 12, 2026
One-click install
npx skills add https://github.com/orthogonal-sh/skills --skill perplexity-orthogonal-sh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: perplexity
Source: https://github.com/orthogonal-sh/skills/tree/main/skills/orthogonal-perplexity
Command: npx skills add https://github.com/orthogonal-sh/skills --skill perplexity-orthogonal-sh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Real-time access to live web data for AI agents powering search and conversational tasks, enabling up-to-date, sourced responses.

Core Features & Use Cases

  • Chat Completions: Generate responses driven by live web data and user prompts.
  • Search: Retrieve ranked results from Perplexity with configurable filtering and domains.
  • Async Tasks: Create and monitor asynchronous chat completion jobs via Sonar APIs for long-running queries.
  • Use Case: Research current events, technical inquiries, or fact-checking with live sources to support decision making.

Quick Start

Ask Perplexity to fetch the latest results for your query and include them in a chat response.

Frequently Asked Questions about perplexity

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

FAQPage Schema
How do I get real-time web search results into AI chat completions?

AI chat completions with real-time web search are generated by sending user prompts to Sonar models, which retrieve live web data and return structured, sourced responses for agent context.

Can I filter web search results by specific domains?

Web search results can be filtered using configurable filters and domain restrictions, allowing you to narrow retrieved data to specific sources before returning ranked outputs to your AI agent.

How do I manage long-running asynchronous web search queries?

Long-running web search queries are managed by creating asynchronous chat completion jobs via Sonar APIs, allowing you to monitor job status and retrieve results once processing is complete.

What parameters can I configure for AI chat with web data?

Configurable parameters for AI chat with web data include model selection, messages, search_mode, and language_preference, enabling customized retrieval and response generation tailored to specific agent contexts.

Does this approach work for fact-checking current events?

Fact-checking current events is supported by retrieving ranked results from live web sources, enabling up-to-date research and verification for technical inquiries and decision-making processes.