Quick Start

Get human QA testing via API.

---

🤖 Are you an AI agent? Read the streamlined docs for AI agents - optimized for machine-readable integration.

---

1. Get Your API Key

Go to the Playground, sign in (or create an account), then grab your API key from the API Keys dashboard.

2. Run Your First Test

Node.js

const response = await fetch('https://runhuman.com/api/run', {
  method: 'POST',
  headers: {
    'Authorization': `Bearer ${process.env.RUNHUMAN_API_KEY}`,
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    url: 'https://example.com/login',
    description: 'Test login: try valid credentials (user@test.com / demo123), then invalid password',
    outputSchema: {
      loginWorks: {
        type: 'boolean',
        description: 'Does login work with valid credentials?'
      },
      errorShown: {
        type: 'boolean',
        description: 'Is error shown for wrong password?'
      }
    }
  })
});

const result = await response.json();
console.log(result.result.data); // { loginWorks: true, errorShown: true }
console.log(`Cost: $${result.costUsd}`); // Cost: $0.216

Python

import requests
import os

response = requests.post(
    'https://runhuman.com/api/run',
    headers={
        'Authorization': f'Bearer {os.environ["RUNHUMAN_API_KEY"]}',
        'Content-Type': 'application/json'
    },
    json={
        'url': 'https://example.com/login',
        'description': 'Test login: try valid credentials (user@test.com / demo123), then invalid password',
        'outputSchema': {
            'loginWorks': {
                'type': 'boolean',
                'description': 'Does login work with valid credentials?'
            },
            'errorShown': {
                'type': 'boolean',
                'description': 'Is error shown for wrong password?'
            }
        }
    }
)

result = response.json()
print(result['result']['data'])  # {'loginWorks': True, 'errorShown': True}
print(f"Cost: ${result['costUsd']}")  # Cost: $0.216

cURL

curl -X POST https://runhuman.com/api/run \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example.com/login",
    "description": "Test login: try valid credentials (user@test.com / demo123), then invalid password",
    "outputSchema": {
      "loginWorks": {
        "type": "boolean",
        "description": "Does login work with valid credentials?"
      },
      "errorShown": {
        "type": "boolean",
        "description": "Is error shown for wrong password?"
      }
    }
  }'

Response (after human completes test):

{
  "status": "completed",
  "result": {
    "success": true,
    "explanation": "Login worked correctly. Error message appeared for invalid password.",
    "data": {
      "loginWorks": true,
      "errorShown": true
    }
  },
  "costUsd": 0.216,
  "testDurationSeconds": 120
}

---

Using MCP with AI Agents

Skip the API - let your agent handle it. See the MCP Guide to get started.

Next: API Reference for full endpoint details.