ORYKSAORYKSADevelopers
Get API keys

Build with an AI employee that already knows the business

One brain for every channel. Call it from your app, your backend or your coding agent, and get replies in the owner's voice, with the business knowledge, rules and plan limits already applied.

curl https://api.oryksa.com/v1/chat \
  -H "Authorization: Bearer oryk_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"message": "Hi! What are your opening hours?"}'
const res = await fetch("https://api.oryksa.com/v1/chat", {
  method: "POST",
  headers: {
    "Authorization": "Bearer " + process.env.ORYKSA_API_KEY,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({ message: "Hi! What are your opening hours?" }),
});
const data = await res.json();
console.log(data.reply, data.conversation_id);
import os, requests

r = requests.post(
    "https://api.oryksa.com/v1/chat",
    headers={"Authorization": f"Bearer {os.environ['ORYKSA_API_KEY']}"},
    json={"message": "Hi! What are your opening hours?"},
    timeout=60,
)
data = r.json()
print(data["reply"], data["conversation_id"])
<?php
$ch = curl_init("https://api.oryksa.com/v1/chat");
curl_setopt_array($ch, [
  CURLOPT_POST => true,
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer " . getenv("ORYKSA_API_KEY"),
    "Content-Type: application/json",
  ],
  CURLOPT_POSTFIELDS => json_encode(["message" => "Hi! What are your opening hours?"]),
]);
$data = json_decode(curl_exec($ch), true);
echo $data["reply"];
import 'dart:convert';
import 'package:http/http.dart' as http;

final res = await http.post(
  Uri.parse('https://api.oryksa.com/v1/chat'),
  headers: {
    'Authorization': 'Bearer $oryksaApiKey',
    'Content-Type': 'application/json',
  },
  body: jsonEncode({'message': 'Hi! What are your opening hours?'}),
);
final data = jsonDecode(res.body);
print(data['reply']);
val body = JSONObject(mapOf("message" to "Hi! What are your opening hours?")).toString()
val req = Request.Builder()
    .url("https://api.oryksa.com/v1/chat")
    .header("Authorization", "Bearer $oryksaApiKey")
    .post(body.toRequestBody("application/json".toMediaType()))
    .build()
OkHttpClient().newCall(req).execute().use { res ->
    val data = JSONObject(res.body!!.string())
    println(data.getString("reply"))
}
var req = URLRequest(url: URL(string: "https://api.oryksa.com/v1/chat")!)
req.httpMethod = "POST"
req.setValue("Bearer \(oryksaApiKey)", forHTTPHeaderField: "Authorization")
req.setValue("application/json", forHTTPHeaderField: "Content-Type")
req.httpBody = try JSONSerialization.data(withJSONObject: ["message": "Hi! What are your opening hours?"])
let (data, _) = try await URLSession.shared.data(for: req)
let json = try JSONSerialization.jsonObject(with: data) as! [String: Any]
print(json["reply"] ?? "")

What you can build

Everything below is live today.

💬

Chat API

Send a customer message, get the AI employee's reply. Conversations keep their context.

Live
🧠

Knowledge (Brain)

Teach pages, FAQs or a whole live site. Keep it in sync from your CMS, shop or ERP.

Live
🔔

Webhooks

Signed events when the AI replies, books an appointment or the plan limit is reached.

Live
🤖

MCP server

Claude Code, Codex, Cursor and VS Code add the AI employee to the site they are building.

Live
🪟

Website chat

One script tag: chat and voice, the business photo, greeting and suggestions.

Live
📱

SDKs for apps

JavaScript, React Native, Flutter, Android, iOS and desktop: the AI employee inside your app.

Live

Start in 3 steps

From zero to your first AI reply in a few minutes.

Create an API key

Account > Developer in the ORYKSA app. The key is shown once.

Teach the business

Send pages or FAQs, or let ORYKSA read the live site.

Send the first message

POST /v1/chat and get the reply in a few seconds.