exa-search

Query web content via Exa endpoints for semantic search and structured insights.

1|Updated Feb 17, 2026
One-click install
npx skills add https://github.com/snqb/my-skills --skill exa-search-snqb
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: exa-search
Source: https://github.com/snqb/my-skills/tree/main/exa-search
Command: npx skills add https://github.com/snqb/my-skills --skill exa-search-snqb

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Enable AI agents to perform comprehensive web searches and retrieve structured insights with optional deep reasoning, citations, and async research capabilities.

Core Features & Use Cases

  • Semantic search: Retrieve meaning-based results beyond simple keyword matching.
  • Deep search with structured output: Generate rich, schema-backed results for complex queries.
  • Find Similar: Discover pages similar to a given URL to broaden research.
  • Contents extraction: Pull text, highlights, and summaries from known URLs.
  • Answer with citations: Obtain direct answers backed by source references.
  • Asynchronous research: Plan and execute long-running, agent-style research tasks.

Quick Start

Set your API key and start a search with the /search endpoint to retrieve results.

Frequently Asked Questions about exa-search

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

FAQPage Schema
How do I get structured web search results with citations for an AI agent?

Structured web search results with citations are retrieved by querying remote web content using the /answer endpoint, which provides direct answers backed by source references. This enables AI agents to return schema-backed outputs.

Can I perform semantic web search instead of simple keyword matching?

Semantic web search is supported to retrieve meaning-based results beyond simple keyword matching. This allows queries to find relevant pages based on conceptual similarity rather than exact text strings.

What is the best way to find pages similar to a specific URL for research?

Finding pages similar to a specific URL is achieved using the /findSimilar endpoint. This discovers related content to broaden research by matching against a provided source link.

How do I extract text and highlights from known URLs via an API?

Text, highlights, and summaries are extracted from known URLs through the /contents endpoint. This allows configurable content handling to pull structured insights directly from target pages.

Does this support asynchronous research for long-running agent tasks?

Asynchronous research is fully supported to plan and execute long-running, agent-style tasks. The /research endpoint enables complex workflows without requiring immediate synchronous responses.

How do I configure deep search to generate rich, schema-backed outputs?

Deep search generates rich, schema-backed results for complex queries using structured output configurations. The /search endpoint handles these requests with optional structured outputs and configurable content handling.