API Documentation - AI Predictions for Sports, Forex & Crypto

Getting Started

The SykikAI API gives you programmatic access to AI-powered predictions across sports, forex, and crypto markets. You can also use our AI chat models for custom analysis and receive predictions via webhooks in real-time.

Base URL

https://sykikai.com/api

Requirements

Developer plan subscription
API key generated from your dashboard

Response Format

All responses are JSON with Content-Type: application/json

Quick test

GET /api/health

curl https://sykikai.com/api/health

Response

{
  "status": "ok",
  "timestamp": "2026-02-11T12:00:00.000000Z"
}

Authentication

Authenticate requests using an API key. Keys are prefixed with sk- and can be created from your dashboard under API Keys.

Two supported methods:

Bearer Token (recommended)

Authorization: Bearer sk-abc123...

Custom Header

X-API-Key: sk-abc123...

Example authenticated request

curl https://sykikai.com/api/predictions \
  -H "Authorization: Bearer sk-your-api-key-here"

Error responses

Status	Cause
401	Missing or invalid API key
403	Your plan does not include API access — upgrade to Developer plan

AI Chat

POST /api/ai/chat

Send a prompt to one of our AI models. Supports optional real-time market data injection for crypto, forex, and commodities analysis.

Request body

Parameter	Type	Required	Description
model	string	Required	Model ID from `/api/models`
prompt	string	Required	Your prompt. Max 10,000 characters
max_tokens	integer	Optional	1–4,000. Default: 1,000
temperature	float	Optional	0–2. Controls randomness. Default: 1.0
include_market_data	boolean	Optional	Inject live market data into prompt context
market_category	string	Optional	`crypto`, `forex`, or `commodities`

POST /api/ai/chat

curl -X POST https://sykikai.com/api/ai/chat \
  -H "Authorization: Bearer sk-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "lightning-pro",
    "prompt": "Analyze the upcoming Liverpool vs Arsenal match",
    "max_tokens": 500,
    "temperature": 0.7
  }'

{
  "response": "Based on current form and head-to-head stats...",
  "model": "Lightning Pro",
  "usage": {
    "input_tokens": 45,
    "output_tokens": 320,
    "total_tokens": 365,
    "cost": "0.01095"
  }
}

Credits

Each request costs (input_tokens + output_tokens) × cost_per_token. Subscription credits are used first, then purchased credits.

Predictions

GET /api/predictions

Returns today's published predictions. Optionally filter by sport.

Query Param	Description
sport	Filter by sport: `football`, `basketball`, `tennis`, etc.

{
  "data": [
    {
      "id": 42,
      "sport": "football",
      "league": "Premier League",
      "home_team": "Liverpool",
      "away_team": "Arsenal",
      "match_time": "2026-02-15T15:00:00Z",
      "market_type": "BTTS",
      "pick": "Yes",
      "odds": 1.72,
      "confidence": "high",
      "analysis": "Both teams have scored in 5 of last 6...",
      "status": "pending",
      "created_at": "2026-02-11T09:00:00Z"
    }
  ],
  "count": 1,
  "date": "2026-02-11"
}

GET /api/predictions/history

Returns settled predictions (won, lost, void) with pagination.

Query Param	Description
per_page	Results per page. Default: 20

GET /api/predictions/{id}

Returns a single published prediction by ID.

Prediction fields

Field	Type	Values
status	string	`pending` `won` `lost` `void` `postponed`
confidence	string	`high` `medium-high` `medium` `value-play`
odds	float	Decimal odds (e.g. 1.85)
match_time	string	ISO 8601 UTC datetime
sport	string	football, basketball, tennis, hockey, mma, boxing, esports, cricket, rugby

Webhooks

Receive predictions in real-time via HTTP POST to your endpoint. When new predictions are published, we send them to all active webhook endpoints on your account. Configure webhooks from your dashboard.

Event prediction.created

Fired when one or more new predictions are published.

Payload

{
  "event": "prediction.created",
  "data": [
    {
      "id": 42,
      "sport": "football",
      "league": "Premier League",
      "home_team": "Liverpool",
      "away_team": "Arsenal",
      "match_time": "2026-02-15T15:00:00Z",
      "market_type": "BTTS",
      "pick": "Yes",
      "odds": 1.72,
      "confidence": "high",
      "analysis": "Both teams scored in 5 of last 6...",
      "status": "pending"
    }
  ],
  "count": 1,
  "timestamp": "2026-02-11T10:30:45Z"
}

Request headers

Header	Description
X-SykikAI-Signature	HMAC-SHA256 signature: `sha256=<hex>`
X-SykikAI-Event	Event type (e.g. `prediction.created`)
X-SykikAI-Delivery	Unique delivery UUID
X-SykikAI-Timestamp	Unix timestamp of delivery

Verifying signatures (PHP)

// Verify the webhook signature
$payload = file_get_contents('php://input');
$signature = hash_hmac('sha256', $payload, $webhookSecret);
$header = str_replace('sha256=', '', $_SERVER['HTTP_X_SYKIKAI_SIGNATURE']);

if (!hash_equals($signature, $header)) {
    http_response_code(401);
    exit('Invalid signature');
}

Retry policy

3 attempts with exponential backoff: 30 seconds, 2 minutes, 10 minutes
Your endpoint must respond with 2xx within 15 seconds
After 10 consecutive failures, the endpoint is automatically disabled

Models & Credits

GET /api/models Public

Returns all available AI models with their pricing. No authentication required.

{
  "models": [
    {
      "name": "Lightning Pro",
      "model": "lightning-pro",
      "cost_per_token": "0.00003000"
    },
    {
      "name": "Swift Basic",
      "model": "swift-basic",
      "cost_per_token": "0.00000200"
    }
  ]
}

How credits work

Subscription credits — included with your plan, reset each billing period
Purchased credits — buy top-ups from the billing page, never expire
Deduction priority — subscription credits are consumed first, then purchased credits
Cost formula — (input_tokens + output_tokens) × cost_per_token

Rate Limits

Endpoints	Limit
Authenticated endpoints (chat, predictions, user)	100 requests / minute
Public endpoints (health, models, market-data)	No limit

Exceeding the rate limit returns 429 Too Many Requests. Rate limits are per API key.

Errors

All errors return a JSON object with an error field.

{
  "error": "Insufficient credits"
}

HTTP status codes

Code	Meaning	When
200	OK	Request succeeded
400	Bad Request	Validation error, insufficient credits
401	Unauthorized	Missing or invalid API key
403	Forbidden	Plan doesn't include this feature
404	Not Found	Resource doesn't exist or isn't published
429	Too Many Requests	Rate limit exceeded (100/min)
500	Server Error	Internal error — contact support

API Reference