Implementation Guide

import Exa from "exa-js";
const exa = new Exa(process.env.EXA_API_KEY!);

const result = await exa.search("Research Acme Corp", {
  type: "deep",
  // Optional: encode rules and judgment criteria
  systemPrompt:
    "You are researching companies for a B2B platform that sells sales intelligence to RevOps teams. The seller's ICP is companies with 50-500 employees that use HubSpot.",
  outputSchema: {
    type: "object",
    required: ["company_name"],
    properties: {
      // Pick any enrichments from the table below
      company_name: {
        type: "string",
        description: "Most commonly-used company name",
      },
      annual_revenue: {
        type: "integer",
        description: "Estimated annual revenue in dollars",
      },
    },
  },
});

console.log(result.output.content);

import Exa from "exa-js";
const exa = new Exa(process.env.EXA_API_KEY!);

const result = await exa.search(
  "Series B fintech companies in the US with 50-200 employees",
  {
    type: "deep",
    // Optional: encode rules and judgment criteria
    systemPrompt:
      "Return up to 10 companies matching the seller's ICP: B2B fintech, recently raised Series B, US-based, 50-200 employees. Score each on icp_fit_score 1-10.",
    outputSchema: {
      type: "object",
      required: ["companies"],
      properties: {
        companies: {
          type: "array",
          items: {
            type: "object",
            required: ["company_name", "domain"],
            properties: {
              company_name: {
                type: "string",
                description: "Most commonly-used company name",
              },
              domain: {
                type: "string",
                description: "Primary company domain, e.g. acme.com",
              },
              headcount: {
                type: "integer",
                description: "Estimated number of employees",
              },
              icp_fit_score: {
                type: "integer",
                description: "ICP fit score from 1 to 10",
              },
              icp_fit_reasoning: {
                type: "string",
                description: "One-sentence rationale for the score",
              },
            },
          },
        },
      },
    },
  }
);

console.log(result.output.content.companies);

import Exa from "exa-js";
const exa = new Exa(process.env.EXA_API_KEY!);

const monitor = await exa.monitors.create({
  name: "Series B Fintech Funding",
  search: {
    query: "New Series B fintech funding rounds in the US",
    numResults: 10,
  },
  trigger: {
    type: "interval",
    period: "1d", // single-unit duration: 1h | 6h | 1d | 7d. Min 1 hour.
  },
  webhook: {
    url: "https://example.com/exa-webhook",
    events: ["monitor.run.completed"],
  },
  // Optional: pass-through context echoed in every webhook delivery
  metadata: {
    slack_channel_id: "C123ABC",
    crm_list_id: "list_abc",
  },
});

// Store this immediately. Only returned on creation; needed to verify
// webhook signatures.
console.log(monitor.webhookSecret);