Quickstart — Moevox API

1. Create an API key

Open the research workspace, go to API keys in your account menu, and create a key. The full key is shown only once — store it somewhere safe.

Keys look like this:

API key format

moe_sk_live_ followed by 43 random characters

2. Check your credits

curl

curl https://api.moevox.com/api/v1/credits \
  -H "Authorization: Bearer moe_sk_live_your_key_here"

Response

{
  "credits_left": 1000
}

3. Submit a report

curl

curl -X POST https://api.moevox.com/api/v1/reports \
  -H "Authorization: Bearer moe_sk_live_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "question": "Which landing page headline converts best?",
    "options": [
      "AI research for product teams",
      "Market validation, automated",
      "Know your market in minutes"
    ],
    "sample_size": 100
  }'

The response contains a request_id. The pipeline runs async and takes a few minutes.

4. Poll for the result

curl

curl https://api.moevox.com/api/v1/reports/req_4f3a9c21e8b7d6c5 \
  -H "Authorization: Bearer moe_sk_live_your_key_here"

Poll every 5 seconds while the questionnaire is being generated, then every 15 seconds during sampling. The status moves through queued → generating_questionnaire → sampling → completed. When completed, the response embeds the full result:

Response — completed

{
  "request_id": "req_4f3a9c21e8b7d6c5",
  "status": "completed",
  "result": {
    "report_url": "https://moevox.com/report/...",
    "title": "Which pricing plan should we launch first?",
    "analysis": { "overview": { "headline": "..." } },
    "sample_data": {
      "questionnaire": { "questions": [] },
      "respondents": []
    },
    "credits_used": 100,
    "credits_left": 900
  }
}

5. Use the report

Open report_url in a browser for the public report page. The analysis JSON contains the structured findings (winner, confidence, segment breakdown, risks, recommendation), and sample_data contains every respondent answer.

6. The whole flow in JavaScript

The same submit-and-poll loop in one snippet (Node.js 18+, no dependencies). Save it as .mjs or wrap the code in anasync function (it uses top-level await):

Node.js

const BASE_URL = 'https://api.moevox.com';
const API_KEY = 'moe_sk_live_your_key_here';

const headers = { Authorization: `Bearer ${API_KEY}` };

async function createReport() {
  const res = await fetch(`${BASE_URL}/api/v1/reports`, {
    method: 'POST',
    headers: { ...headers, 'Content-Type': 'application/json' },
    body: JSON.stringify({
      question: 'Which pricing plan should we launch first?',
      options: ['Monthly $29', 'Monthly $49', 'Annual $299'],
      sample_size: 100,
    }),
  });
  if (!res.ok) throw new Error(`HTTP ${res.status}: ${await res.text()}`);
  return res.json(); // { request_id, status: 'queued', credits_estimated, credits_left }
}

async function pollUntilDone(requestId) {
  for (;;) {
    const res = await fetch(`${BASE_URL}/api/v1/reports/${requestId}`, { headers });
    const data = await res.json();
    if (data.status === 'completed' || data.status === 'failed') return data;
    await new Promise((r) => setTimeout(r, 5000));
  }
}

const { request_id } = await createReport();
const report = await pollUntilDone(request_id);
console.log(report.result); // report_url, analysis, sample_data, credits_used, credits_left

This mirrors the curl flow above. For a polling loop that also handles credits and errors, copy the AI Prompt.