Text Analysis API Usage, Sentiment & Content Moderation

Authentication

All requests must be signed using your apiSecret . Send the generated signature in header x-signature .

Required fields in request body:

• • projectId
• • apiKey
• • nonce (random alphanumeric exactly 10 characters)
• • timestamp
• • text (for single request)
• • texts (for batch request)

⚠️ All fields (including text/texts) must be included while generating the signature.

Signature Generation

1. Send a POST request with application/json body
2. Include required fields + text (single) or texts (batch)
3. Create payload including ALL fields
4. Sort keys before stringify
5. Generate HMAC-SHA256 using apiSecret
6. Send in x-signature header

import crypto from "node:crypto";

const payload = {
  nonce,
  projectId,
  apiKey,
  timestamp,
  ...(isBatch ? { texts } : { text })
};

const stable = JSON.stringify(
  Object.keys(payload)
    .sort()
    .reduce((a, k) => {
      a[k] = payload[k];
      return a;
    }, {})
);

const signature = crypto
  .createHmac("sha256", apiSecret)
  .update(stable)
  .digest("hex");

const payload = {
  nonce,
  projectId,
  apiKey,
  timestamp,
  ...(isBatch ? { texts } : { text })
};

const stable = JSON.stringify(
  Object.keys(payload)
    .sort()
    .reduce((a, k) => {
      a[k] = payload[k];
      return a;
    }, {})
);

import hmac, hashlib, json

payload = {
  "nonce": nonce,
  "projectId": projectId,
  "apiKey": apiKey,
  "timestamp": timestamp,
  **({"texts": texts} if is_batch else {"text": text})
}

stable = json.dumps(payload, sort_keys=True)

signature = hmac.new(
  apiSecret.encode(),
  stable.encode(),
  hashlib.sha256
).hexdigest()

⚠️ text/texts MUST be included in signature calculation, otherwise request will fail.

Single Text

Analyze a single text input. Max length: 500 characters .

POST /v1/text/analyze

{
  "text": "I loved the service!",
  "projectId": "...",
  "apiKey": "...",
  "nonce": "...",
  "timestamp": 123456
}

Batch (Max 10 Texts)

Send multiple texts. Combined length must be ≤ 500 characters .

POST /v1/text/analyze/batch

{
  "texts": ["text1", "text2"],
  "projectId": "...",
  "apiKey": "...",
  "nonce": "...",
  "timestamp": 123456
}

Response Format

Each request returns probability scores (0 → 1) for different content signals. Higher value = stronger confidence.

{
  "success": true,
  "data": {
    "toxic": 0.0087,
    "obsceneContent": 0.0217,
    "threateningContent": 0.0182,
    "insultingLanguage": 0.0755,
    "identityAttack": 0.1063,
    "sexuallyExplicit": 0.0349,
    "spam": 1,
    "phishing": 0,
    "sentiment": 0.3579
  }
}