Add AI Chat to Your App With One API Call

TL;DR: The aiCHATS public Bot API gives developers the same chat engine that powers the social channels, over a REST call. You send a message and a user id, you get an AI reply, and the server handles conversation state, history, and context for you.

Why an API, not another widget

The website widget and the social channels cover most businesses. A developer building a product needs something else: programmatic access. The aiCHATS Bot API exposes the engine behind the aiCHATS platform so you can put AI chat inside your own app, portal, or internal tool. Our automation service helps wire it into a custom build.

The point is what you do not have to build. No prompt engineering, no conversation-state machine, no context-window juggling, no model provider config, no token-limit handling. You call an endpoint and the answer comes back.

What the API exposes

The surface is small on purpose:

• Bots. Create and manage AI bots, each with its own business description, knowledge base, tone, language, and greeting or fallback messages.
• Chat. Send a message and get a reply, as a single JSON response or a streamed one. Pass a user id and the server resumes that user's conversation, or pass a conversation id to be explicit.
• Conversations. Create, list, fetch with history, or delete. The server stores and prunes message history for you.
• Usage. Read messages used against your limit, tokens consumed, and the current period.
• Keys. Generate and rotate API keys, shown once and revocable.

The server holds the conversation

The hard part of a chat feature is memory. A naive integration has to store every message, decide how much to send back to the model, and trim the context before it overflows. The aiCHATS API does that for you. Send a user id with each message and the server threads the conversation, keeping the recent history that matters. The same memory model the channels use is described in conversation memory tiers.

Streaming or single response

For a chat interface, replies that stream in word by word feel alive. The API supports a streamed mode over the same endpoint, the way the website widget streams, covered in streaming AI chat on your site. For a backend job that needs the whole answer at once, the single JSON response fits better. You pick per call.

One quota, one engine

API messages draw from the same subscription pool as your social channels. There is no separate meter for API versus Messenger, which keeps billing simple, the same single-quota model explained in one message quota across all channels. Rate limits scale with your plan, so a higher tier gets more requests per minute.

Who reaches for the API

Three kinds of builders use it. An agency that wants to offer white-label AI chat to its own clients. A SaaS product that needs a support assistant inside the app. A business with a custom portal that wants the same brain its customers meet on Instagram. In each case the API is treated as one more channel for the same AI Brain, the model described in one AI Brain, five channels.

Keys and safety

Each request is authenticated with an API key, shown once at creation, stored as a hash, and revocable if it leaks. You can rotate a key without downtime. Because the engine runs server-side, your application never holds a model provider key or prompt logic, which keeps your own surface small. For a wider view of how aiCHATS compares to other chatbot stacks, see the chatbot platforms comparison.

Where to start

Create a bot, generate a key, and send your first message with a user id. The reply, the history, and the usage tracking come back without any extra plumbing on your side. When you want a custom integration built around it, start with our automation service, and for the business case read the complete chatbot guide for business.

Related reading

• aiCHATS: One AI Brain Across Every Channel
• Streaming AI Chat on Your Site
• Add an AI Sales Rep With One Line of Code
• One Message Quota Across All Channels

FAQ

Do I need to manage conversation history myself?

No. Pass a user id with each message and the server stores and prunes the history, resuming the right conversation for you.

Can the API stream replies?

Yes. The chat endpoint supports a streamed mode for live interfaces and a single JSON response for backend use.

Is the API metered separately from my channels?

No. API messages draw from the same subscription quota as Messenger, Instagram, WhatsApp, Telegram, and the widget.

How is a request authenticated?

With an API key shown once at creation, stored as a hash, and revocable. You can rotate a key without downtime.