> ## Documentation Index
> Fetch the complete documentation index at: https://docs.chatzy.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Overview

Interact with the conversational AI agent using this **3-step process**:

### 1. **Start Conversation** (`/init_conversation`)

* **Purpose**: Create a new conversation session.
* **Required**:
  * `chatbot_id` -> Get this from **Conversational AI Agents** > open your agent's menu (⋮ / hamburger) > **Copy Agent ID**.
* **Response**: Returns a `conversation_id` (used in later steps) and a intorductory message (if enabled).

### 2. **Send Message** (`/get_inference`)

* **Purpose**: Send a user message to the AI.
* **Required**:
  * `chatbot_id` (from step 1)
  * `conversation_id` (from `/init_conversation`)
  * `message` (your text)
* **Response**: Sends user message for processing and returns a `temp_token`.

### 3. **Stream Reply** (`/get-response`)

* **Purpose**: Get the AI's response in real time via SSE (Server-Sent Events).
* **Required**:
  * `temp_token` (from `/get_inference` response or generated separately)
* **Response**: Streams reply response chunks and control signals until the answer is complete.

> **⚠️ Important**: This flow must be repeated for each new user message or session. The `temp_token` expires quickly (validity of `1 minute`) and cannot be reused.