API Documentation — AuthentiText AI

Base URL

https://your-app.onrender.com

Replace with your deployed Render URL

GET /

Health check endpoint. Returns service status.

Response

{
  "status": "online",
  "service": "AuthentiText AI API",
  "version": "1.0.0",
  "endpoints": {
    "POST /predict": "Analyze text for AI generation probability"
  }
}

POST /predict

Analyze text and predict whether it is AI-generated or human-written.

Request Headers

Content-Type: application/json

Request Body

{
  "text": "Your text to analyze (minimum 50 words)"
}

✅ Success Response (200)

{
  "prediction": "AI Generated",
  "confidence": 0.8733,
  "label": 1,
  "word_count": 70,
  "human_probability": 0.1267,
  "ai_probability": 0.8733
}

Response Fields

Field	Type	Description
prediction	string	"AI Generated" or "Human Written"
confidence	float	Confidence score (0 to 1)
label	int	0 = Human, 1 = AI
word_count	int	Number of words in input
human_probability	float	Probability of being human-written
ai_probability	float	Probability of being AI-generated

❌ Error Responses

400 — { "error": "Text too short. Minimum 50 words required. You provided 12 words." }

400 — { "error": "Text field is empty. Please provide text to analyze." }

400 — { "error": "Invalid request. Please send JSON with a 'text' field." }

📎 cURL Example

curl -X POST https://your-app.onrender.com/predict \
  -H "Content-Type: application/json" \
  -d '{"text": "Your text to analyze goes here..."}'

📎 JavaScript (fetch) Example

const response = await fetch('/predict', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({ text: 'Your text here...' })
});
const data = await response.json();
console.log(data.prediction, data.confidence);