Exa SDK Patterns

Overview

Production-ready patterns for the exa-js SDK. Covers client singletons, typed wrappers, error handling, retry logic, and response validation for real Exa API methods.

Prerequisites

• exa-js installed and EXA_API_KEY configured
• TypeScript project with strict mode
• Familiarity with async/await and error handling

Instructions

Step 1: Client Singleton

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

let instance: Exa | null = null;

export function getExa(): Exa {
  if (!instance) {
    const apiKey = process.env.EXA_API_KEY;
    if (!apiKey) {
      throw new Error("EXA_API_KEY not set. Get one at https://dashboard.exa.ai");
    }
    instance = new Exa(apiKey);
  }
  return instance;
}

Step 2: Typed Search Wrapper

// src/exa/search.ts
import Exa from "exa-js";
import { getExa } from "./client";

interface ExaSearchOptions {
  type?: "auto" | "neural" | "keyword" | "fast" | "instant" | "deep" | "deep-reasoning";
  numResults?: number;
  includeDomains?: string[];
  excludeDomains?: string[];
  startPublishedDate?: string;
  endPublishedDate?: string;
  category?: "company" | "research paper" | "news" | "tweet" | "personal site" | "financial report" | "people";
  includeText?: string[];
  excludeText?: string[];
}

interface ExaContentsOptions {
  text?: boolean | { maxCharacters?: number; includeHtmlTags?: boolean };
  highlights?: boolean | { maxCharacters?: number; query?: string };
  summary?: boolean | { query?: string };
  livecrawl?: "always" | "preferred" | "fallback" | "never";
  livecrawlTimeout?: number;
  subpages?: number;
  subpageTarget?: string | string[];
}

export async function exaSearch(query: string, opts: ExaSearchOptions = {}) {
  const exa = getExa();
  return exa.search(query, {
    type: opts.type ?? "auto",
    numResults: opts.numResults ?? 10,
    ...opts,
  });
}

export async function exaSearchWithContents(
  query: string,
  searchOpts: ExaSearchOptions = {},
  contentOpts: ExaContentsOptions = {}
) {
  const exa = getExa();
  return exa.searchAndContents(query, {
    type: searchOpts.type ?? "auto",
    numResults: searchOpts.numResults ?? 10,
    ...searchOpts,
    ...contentOpts,
  });
}

Step 3: Error Handling Wrapper

// src/exa/safe.ts
interface ExaResult<T> {
  data: T | null;
  error: ExaError | null;
}

interface ExaError {
  status: number;
  message: string;
  tag?: string;
  requestId?: string;
  retryable: boolean;
}

function classifyError(err: any): ExaError {
  const status = err.status || err.response?.status || 500;
  const retryable = status === 429 || status >= 500;
  return {
    status,
    message: err.message || "Unknown error",
    tag: err.error_tag || err.tag,
    requestId: err.requestId || err.request_id,
    retryable,
  };
}

export async function safeExaCall<T>(
  operation: () => Promise<T>
): Promise<ExaResult<T>> {
  try {
    const data = await operation();
    return { data, error: null };
  } catch (err: any) {
    const error = classifyError(err);
    console.error(`[Exa Error] ${error.status}: ${error.message}`, {
      tag: error.tag,
      requestId: error.requestId,
      retryable: error.retryable,
    });
    return { data: null, error };
  }
}

// Usage:
// const { data, error } = await safeExaCall(() =>
//   exa.searchAndContents("query", { numResults: 5, text: true })
// );

Step 4: Retry with Exponential Backoff

// src/exa/retry.ts
export async function withRetry<T>(
  operation: () => Promise<T>,
  config = { maxRetries: 3, baseDelayMs: 1000, maxDelayMs: 30000 }
): Promise<T> {
  for (let attempt = 0; attempt <= config.maxRetries; attempt++) {
    try {
      return await operation();
    } catch (err: any) {
      const status = err.status || err.response?.status || 0;

      // Only retry on rate limits (429) and server errors (5xx)
      if (status !== 429 && (status < 500 || status >= 600)) throw err;
      if (attempt === config.maxRetries) throw err;

      const delay = Math.min(
        config.baseDelayMs * Math.pow(2, attempt) + Math.random() * 500,
        config.maxDelayMs
      );
      console.log(`[Exa] Retry ${attempt + 1}/${config.maxRetries} in ${delay.toFixed(0)}ms`);
      await new Promise(r => setTimeout(r, delay));
    }
  }
  throw new Error("Unreachable");
}

// Usage:
// const results = await withRetry(() =>
//   exa.searchAndContents("query", { numResults: 5, text: true })
// );

Step 5: Response Validation with Zod

// src/exa/validate.ts
import { z } from "zod";

const ExaResultSchema = z.object({
  url: z.string().url(),
  title: z.string().nullable(),
  score: z.number(),
  publishedDate: z.string().nullable().optional(),
  text: z.string().optional(),
  highlights: z.array(z.string()).optional(),
  summary: z.string().optional(),
});

const ExaSearchResponseSchema = z.object({
  results: z.array(ExaResultSchema),
  autopromptString: z.string().optional(),
});

export function validateSearchResponse(response: unknown) {
  return ExaSearchResponseSchema.parse(response);
}

Error Handling

Pattern	Use Case	Benefit
Singleton	All API calls	Single client instance, consistent config
Safe wrapper	Non-critical searches	Prevents uncaught exceptions
Retry logic	Rate limits and 5xx	Automatic recovery from transient failures
Zod validation	Response processing	Catches unexpected API response changes
Typed options	IDE support	Autocomplete and compile-time checks

Examples

Factory Pattern (Multi-tenant)

const clients = new Map<string, Exa>();

export function getExaForTenant(tenantId: string): Exa {
  if (!clients.has(tenantId)) {
    const apiKey = getTenantApiKey(tenantId); // from your config/vault
    clients.set(tenantId, new Exa(apiKey));
  }
  return clients.get(tenantId)!;
}

Combined: Safe + Retry + Typed

async function resilientSearch(query: string) {
  return safeExaCall(() =>
    withRetry(() =>
      exaSearchWithContents(
        query,
        { type: "neural", numResults: 5 },
        { text: { maxCharacters: 2000 }, highlights: true }
      )
    )
  );
}

Resources

• exa-js TypeScript SDK
• Exa Error Codes
• Zod Documentation

Next Steps

Apply patterns in exa-core-workflow-a for real-world search usage.