exa

Find similar web content and retrieve page data via Exa neural search.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Neural content discovery can be time-consuming and noisy; Exa provides a neural web search interface to find similar content, surface relevant sources, and retrieve page data for deeper analysis.

Core Features & Use Cases

  • Neural similarity search: Find pages and resources similar to a given URL or query and optionally fetch full contents.
  • Content retrieval & summaries: Retrieve page text, metadata, and concise summaries for quick evaluation.
  • Use Case: Use Exa to perform competitive research by collecting related articles and their citations, then synthesize a literature review.

Quick Start

Run a sample Exa search to explore related content.

Frequently Asked Questions about exa

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

FAQPage Schema
How do I find similar content and related articles using neural web search?

Neural web search finds similar content by querying a given URL or text to surface relevant sources and retrieve page data. It uses neural similarity matching to locate related articles and resources for deeper analysis.

Can I retrieve full page contents and summaries for competitive research?

Yes, you can retrieve full page contents, metadata, highlights, and concise summaries for competitive research. The search returns page text and sourced summaries to enable quick evaluation of related articles and their citations.

What is the best way to perform market research and source citations automatically?

The best way to perform market research is using neural search to collect related articles and their citations. You can synthesize a literature review by fetching page contents and generating sourced summaries from the discovered data.

Does neural search support asynchronous tasks for finding similar links?

Yes, neural search supports creating asynchronous tasks to find similar links and retrieve page contents. This allows you to run queries, fetch full text, and return results with optional highlights and summaries without blocking execution.

What are the limitations of using neural content discovery for fact-finding?

Neural content discovery for fact-finding can return noisy or time-consuming results if queries are too broad. While it retrieves page text and metadata, the quality of sourced summaries depends on the relevance of the initially surfaced content.