search

Orchestrate web and dataset searches across multiple backends with parallel API calls.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Web search and data gathering across multiple platforms can be fragmented, requiring users to manually select and combine results from various APIs.

Core Features & Use Cases

  • Multi-backend search: automatically choose among Tavily, Exa, Andi, Linkup, Valyu, and SearchAPI to fetch relevant results.
  • Research & verification: gather, compare, and cite information from diverse sources for deep research or quick answers.
  • Platform-specific lookups: perform targeted queries on platforms like YouTube, Amazon, eBay, and news outlets.
  • Combine results: blend data from multiple backends for comprehensive insights.

Quick Start

Ask the AI to perform a Tavily search for the latest AI agents and return a summarized set of results with citations.

Frequently Asked Questions about search

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

FAQPage Schema
How do I search across multiple web APIs and combine the results?

To search across multiple web APIs, this skill orchestrates parallel calls to Tavily, Exa, Andi, Linkup, Valyu, and SearchAPI. It harmonizes inputs, ranks the combined results, and extracts citations for structured consumption.

Can I perform platform-specific lookups on YouTube or Amazon using a web search API?

Yes, platform-specific lookups on YouTube, Amazon, eBay, and news outlets are supported. The skill routes your query to the appropriate backend to fetch targeted platform data and blends it with general web results.

What is the best way to gather citations and verify information for deep research?

The best way to gather citations for deep research is using the multi-backend routing mechanism. It fetches data from diverse sources, compares the information, and optionally extracts citations for structured verification.

How does multi-API search routing work for general web queries?

Multi-API search routing works by automatically selecting among Tavily, Exa, Andi, Linkup, Valyu, and SearchAPI. It executes parallel API calls based on your general web query, ranks the aggregated results, and returns a unified dataset.

Do I need to manually select a specific search backend to fetch relevant results?

No, you do not need to manually select a search backend. The skill automatically chooses among Tavily, Exa, Andi, Linkup, Valyu, and SearchAPI to fetch relevant results based on input harmonization and query type.

Why combine data from multiple search backends instead of using a single API?

Combining data from multiple search backends prevents fragmented results and provides comprehensive insights. The skill ranks blended data from diverse sources, ensuring deeper research coverage than a single API could offer.