polymer-pay-ortho-exa

name	polymer-pay-ortho-exa
description	USE THIS SKILL WHEN: the user wants AI-powered web search with semantic understanding, neural search, content extraction, and research capabilities. Provides AI search via Exa through the Polymer Pay proxy.
endpoints	[{"path":"/search","method":"POST","price":"$0.01","description":"Neural web search"},{"path":"/answer","method":"POST","price":"$0.01","description":"LLM answers with citations"},{"path":"/findSimilar","method":"POST","price":"$0.01","description":"Find similar content"},{"path":"/contents","method":"POST","price":"$0.01","description":"Extract URL contents"},{"path":"/research/v1","method":"POST","price":"$0.01","description":"Async research tasks"},{"path":"/research/v1","method":"GET","price":"$0.01","description":"List research tasks"},{"path":"/research/v1/{id}","method":"GET","price":"$0.01","description":"Get research results"}]
metadata	{"polymer-pay-skill":{"emoji":"🧠","requires":{"env":["POLYMER_PAY_API_KEY"]},"primaryEnv":"POLYMER_PAY_API_KEY"}}
registries	{}
provider	orthogonal

Exa (x402.orth.sh)

Exa is a search engine made for AIs. It provides semantic web search, neural embeddings-based search, content extraction, and AI-powered research with citations through the Polymer Pay proxy via x402.orth.sh.

Authentication

All requests route through the Polymer Pay proxy. Include your Polymer Pay API key in every request:

{
  "headers": {
    "Content-Type": "application/json",
    "x-polymer-pay-api-key": "{{POLYMER_PAY_API_KEY}}"
  }
}

Base URL: https://pay.polymerlabs.org/proxy/https/x402.orth.sh/exa

Common Operations

Search

Intelligent web search with neural embeddings and semantic understanding.

Pricing: $0.01

{
  "method": "POST",
  "url": "https://pay.polymerlabs.org/proxy/https/x402.orth.sh/exa/search",
  "headers": {
    "Content-Type": "application/json",
    "x-polymer-pay-api-key": "{{POLYMER_PAY_API_KEY}}"
  },
  "body": {
    "query": "latest developments in quantum computing",
    "numResults": 10,
    "type": "neural",
    "text": true
  }
}

Response: Search results with URLs, titles, snippets, and full text content.

Answer

Get an LLM answer to a question informed by Exa search results.

Pricing: $0.01

{
  "method": "POST",
  "url": "https://pay.polymerlabs.org/proxy/https/x402.orth.sh/exa/answer",
  "headers": {
    "Content-Type": "application/json",
    "x-polymer-pay-api-key": "{{POLYMER_PAY_API_KEY}}"
  },
  "body": {
    "query": "What is the capital of France?",
    "text": true,
    "stream": false
  }
}

Response: Generated answer with source citations and reference URLs.

Find Similar

Find similar links to a given URL.

Pricing: $0.01

{
  "method": "POST",
  "url": "https://pay.polymerlabs.org/proxy/https/x402.orth.sh/exa/findSimilar",
  "headers": {
    "Content-Type": "application/json",
    "x-polymer-pay-api-key": "{{POLYMER_PAY_API_KEY}}"
  },
  "body": {
    "url": "https://example.com/article",
    "numResults": 10,
    "text": true
  }
}

Get Contents

Extract full page contents, summaries, and metadata from URLs.

Pricing: $0.01

{
  "method": "POST",
  "url": "https://pay.polymerlabs.org/proxy/https/x402.orth.sh/exa/contents",
  "headers": {
    "Content-Type": "application/json",
    "x-polymer-pay-api-key": "{{POLYMER_PAY_API_KEY}}"
  },
  "body": {
    "urls": ["https://example.com/article1", "https://example.com/article2"],
    "text": true,
    "summary": true,
    "highlights": true
  }
}

Research

Create an asynchronous research task that explores the web and synthesizes findings.

Pricing: $0.01

{
  "method": "POST",
  "url": "https://pay.polymerlabs.org/proxy/https/x402.orth.sh/exa/research/v1",
  "headers": {
    "Content-Type": "application/json",
    "x-polymer-pay-api-key": "{{POLYMER_PAY_API_KEY}}"
  },
  "body": {
    "instructions": "Research the current state of AI regulation in the EU and US",
    "model": "exa-research"
  }
}

Search Types

• neural — Embeddings-based semantic search (best quality)
• auto — Intelligently combines neural and other methods (default)
• fast — Streamlined search for speed
• deep — Comprehensive search with query expansion

When to Use

• Semantic search — Find conceptually related content, not just keyword matches
• AI answers — Get direct answers with source citations
• Content extraction — Extract full text, summaries, and highlights from URLs
• Research automation — Create async research tasks with synthesized reports
• Similar content discovery — Find related articles and resources
• RAG applications — Use context mode to combine results into LLM-ready strings

Best Practices

• Choose search type — Use "neural" for quality, "fast" for speed, "deep" for comprehensive results
• Enable text extraction — Set text: true to get full page content for analysis
• Use highlights — Extract the most relevant text snippets from pages
• Include summaries — Get AI-generated summaries of page content
• Domain filtering — Use includeDomains/excludeDomains to control sources
• Date filtering — Use startPublishedDate/endPublishedDate for recent content
• For errors — See @skills/polymer-pay-api-errors/SKILL.md for complete error code reference and troubleshooting