Help center
Your AI agent reads photos, voice notes and documents on WhatsApp
The TBit agent understands what your customer sends: it identifies products in photos, reads payment receipts, transcribes voice notes and processes PDFs. What to expect and what not to.
5 min read

What it does
The agent understands what your customer sends, not just text: it reads images, listens to voice notes and processes documents.
How it works
Photos: the agent sees the content. If the customer sends a photo of a product, it can identify it and look it up in your catalog; if they send a payment receipt, it reads the amount and date.
Voice notes: they get transcribed and the agent replies as if they were text.
Documents (PDF and similar): the agent extracts the text and replies about the content.
Sending media: the agent also sends product photos and business documents when the conversation calls for it.
What to expect and what not to
It works really well with clear content: legible receipts, sharp photos, audios without much noise. With a blurry photo or a cut-off audio, the agent asks the customer to resend it or type it out. It doesn't guess what it can't see.
Where to set it up
It needs no setup: it comes on. The files that come in and go out stay in the conversation, visible in the Chats inbox.
What NOT to put in the prompt
Instructions like "if they send you an image, describe it" or "transcribe the audios". The engine already does it. The prompt can define what to do with the result, for example "when a receipt comes in, confirm the order and let your team know".
Train it with your own AI
Do you use ChatGPT or Claude? Connect them to your TBit agent and let them help you train it with your real data: on the platform go to Agent settings, Connect your AI section, generate your token and add the connector https://api.tbit.app/mcp. Your assistant can review the prompt, the memories and recent conversations, and suggest improvements.