API Documentation

How it works

Every agent you create gets its own ID (looks like agt_xxxxxxxx, shown on the agent's page). You send it a message with your API key, and it replies using only the knowledge you gave that agent.

Authentication

Every request needs your API key in the Authorization header. Get a key from your dashboard.

Authorization: Bearer YOUR_API_KEY

Base URL

https://ai.fireclashpro.com

Send a message to an agent

POST /v1/agents/{agent_id}/chat

curl https://ai.fireclashpro.com/v1/agents/agt_yourid/chat \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"message": "What products do you offer?"}'

Response:

{
  "agent_id": "agt_yourid",
  "reply": "We offer handmade candles and soaps..."
}

Code examples

Try it yourself with the in-browser API Tester, or use one of these:

curl https://ai.fireclashpro.com/v1/agents/agt_yourid/chat \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"message": "What products do you offer?"}'

import requests

response = requests.post(
    "https://ai.fireclashpro.com/v1/agents/agt_yourid/chat",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={"message": "What products do you offer?"},
)
print(response.json()["reply"])

const res = await fetch("https://ai.fireclashpro.com/v1/agents/agt_yourid/chat", {
  method: "POST",
  headers: {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({ message: "What products do you offer?" }),
});
const data = await res.json();
console.log(data.reply);

$ch = curl_init("https://ai.fireclashpro.com/v1/agents/agt_yourid/chat");
curl_setopt_array($ch, [
    CURLOPT_RETURNTRANSFER => true,
    CURLOPT_POST => true,
    CURLOPT_HTTPHEADER => [
        "Authorization: Bearer YOUR_API_KEY",
        "Content-Type: application/json",
    ],
    CURLOPT_POSTFIELDS => json_encode(["message" => "What products do you offer?"]),
]);
$response = json_decode(curl_exec($ch), true);
echo $response["reply"];

HttpClient client = HttpClient.newHttpClient();
String body = "{\"message\": \"What products do you offer?\"}";

HttpRequest request = HttpRequest.newBuilder()
    .uri(URI.create("https://ai.fireclashpro.com/v1/agents/agt_yourid/chat"))
    .header("Authorization", "Bearer YOUR_API_KEY")
    .header("Content-Type", "application/json")
    .POST(HttpRequest.BodyPublishers.ofString(body))
    .build();

HttpResponse<String> response = client.send(request, HttpResponse.BodyHandlers.ofString());
System.out.println(response.body());

val client = OkHttpClient()
val mediaType = "application/json".toMediaType()
val body = """{"message": "What products do you offer?"}""".toRequestBody(mediaType)

val request = Request.Builder()
    .url("https://ai.fireclashpro.com/v1/agents/agt_yourid/chat")
    .addHeader("Authorization", "Bearer YOUR_API_KEY")
    .post(body)
    .build()

client.newCall(request).execute().use { response ->
    println(response.body?.string())
}

Agent status

An agent must have at least one knowledge source added before it can chat — check its status on the agents page. Calling an agent that isn't ready yet returns 400.

Rate limits

Each API key has a monthly request limit based on your plan (see pricing). Going over returns HTTP 429.

Errors

401	Missing, invalid, or revoked API key
404	Agent not found (wrong ID, or it belongs to a different account)
400	Agent exists but isn't ready yet (no knowledge added)
429	Request limit reached for this key
502	Backend model router unreachable — try again shortly