exa-reference-architecture

name	exa-reference-architecture
description	Implement Exa reference architecture for search pipelines, RAG, and content discovery. Use when designing new Exa integrations, reviewing project structure, or establishing architecture standards for neural search applications. Trigger with phrases like "exa architecture", "exa project structure", "exa RAG pipeline", "exa reference design", "exa search pipeline".
allowed-tools	Read, Grep
version	1.11.0
license	MIT
author	Jeremy Longshore <[e-mail]>
tags	["saas","exa","architecture","rag"]
compatibility	Designed for Claude Code, also compatible with Codex and OpenClaw

Exa Reference Architecture

Overview

Production architecture for Exa neural search integration. Covers search service design, content extraction pipeline, RAG integration, domain-scoped search profiles, and caching strategy.

Architecture Diagram

┌──────────────────────────────────────────────────────────┐
│                  Application Layer                        │
│   RAG Pipeline  |  Research Agent  |  Content Discovery   │
└──────────┬──────────────┬───────────────┬────────────────┘
           │              │               │
           ▼              ▼               ▼
┌──────────────────────────────────────────────────────────┐
│                Exa Search Service Layer                    │
│  ┌────────────┐  ┌────────────┐  ┌──────────────────┐    │
│  │ search()   │  │ findSimilar│  │ getContents()    │    │
│  │ neural/    │  │ (URL seed) │  │ (known URLs)     │    │
│  │ keyword/   │  └────────────┘  └──────────────────┘    │
│  │ auto/fast  │                                           │
│  └────────────┘                  ┌──────────────────┐    │
│                                  │ answer() /       │    │
│  Content Options:                │ streamAnswer()   │    │
│  text | highlights | summary     └──────────────────┘    │
│                                                           │
│  ┌────────────────────────────────────────────────────┐  │
│  │              Result Cache (LRU + Redis)             │  │
│  └────────────────────────────────────────────────────┘  │
└──────────────────────────────────────────────────────────┘
         │
         ▼
┌──────────────────────────────────────────────────────────┐
│  api.exa.ai — Exa Neural Search API                      │
│  Auth: x-api-key header | Rate: 10 QPS default           │
└──────────────────────────────────────────────────────────┘

Instructions

Step 1: Search Service Layer

// src/exa/service.ts
import Exa from "exa-js";

const exa = new Exa(process.env.EXA_API_KEY);

interface SearchRequest {
  query: string;
  type?: "auto" | "neural" | "keyword" | "fast" | ;
  ?: ;
  ?: ;
  ?: ;
  ?: [];
  ?: [];
  ?:  |  |  |  | ;
}

  {
  ?:  | { ?:  };
  ?:  | { ?: ; ?:  };
  ?:  | { ?:  };
}

   () {
   exa.(req., {
    : req. || ,
    : req. || ,
    : req.,
    : req.,
    : req.,
    : req.,
    : req.,
    ...content,
  });
}

   () {
   exa.(url, {
    numResults,
    : { :  },
    : ,
  });
}