What a Voice Agent Knows
Where a phone agent finds its answers — the brand's own knowledge — and what else it can do while a caller is on the line, including running your workflows and API tools mid-call.
Where the Answers Come From
A voice agent answers from its brand's knowledge. Everything on Brand → (your brand) → Knowledge — knowledge bases, documentation sets, uploaded files, pasted text snippets, question-and-answer entries and pages of your website we read for you — plus the brand's product catalog is searched in one pass on every lookup. It is exactly the same material the brand's chat AI reads.
There is no separate "voice knowledge base" to maintain: the phone agent and the chat AI share the same brand, not the same bot. A brand does not need a chat AI at all for its phone agent to answer from that knowledge.
The agent also follows the brand's company description and the Answer instructions of the brand's AI agent (without an AI agent, the brand's details), together with its own instructions, and it knows its own name.
What the Agent Can Do on a Call
Look something up
Searches the brand's whole body of knowledge and its products in one pass. This is the default for almost every factual question — the agent tries it whenever a question could plausibly be about your company.
Browse your catalog
Ask the catalog for its own structure — your categories, and the products in one of them as names with live prices, in one order called most popular (products with a priority first, then by popularity, out-of-stock last). It can also filter (a price limit, in stock, on sale, a flavor or size, a word such as vegan) and, when the caller asks, sort by highest rated, newest or cheapest from your own data. Offered whenever the agent's brand has at least one active product, with no Yaplet required.
End the call
When the conversation is finished the agent can hang up by itself. Every agent has this, on every brand.
Run a workflow or call an API tool
Fetching live data from your own systems — an order status, stock, a booking — or running one of your scripted chat workflows that can run on a phone call. Offered whenever the brand has any, with no Yaplet chat AI required; see the next section.
Tell your team
Only for an agent with When a caller asks for a person filled in: when it agrees something your team has to do, such as a callback, it alerts your team and the conversation waits in the inbox. Without that text the agent has no way to tell your team.
Talking About Products on the Phone
A caller asking "what do you sell?" or "which proteins do you have?" gets the same answer your chat AI would give, built from your real categories and the same most popular order rather than from whichever products a search happened to find.
Two things are adjusted for the fact that nobody is looking at a screen:
- No ids and no links are ever spoken. The agent works with names only.
- Long lists are not read aloud. The agent names only a few products at once, and asks what the caller needs rather than reciting a category. A question for your best or top products gets the most popular order.
Product conduct rules like the ones that govern chat apply on a call — no invented quality tiers, no popularity claims beyond the most popular order, no describing a product beyond what its sheet states.
Without Yaplet on the Brand
Calls handled by an agent whose brand has no Yaplet are saved completely normally — transcript, inbox conversation, recording, minutes billed. The one thing they do not produce is AI answer-quality reporting, because those figures are counted per chat bot and there is no bot to count them against.
API Tools Are Your Own Endpoints
"API tools" are not built-in helpers that read the caller's number or the time of day. They are endpoints you define yourself, which the agent calls mid-conversation to fetch something that changes.
You add one with Add knowledge → API tool on Brand → (your brand) → Knowledge. Your tools then sit in the API tools group on that page, each with its name, its method and the start of its description, and the brand's chat AI uses the same ones. In the tool editor the box you describe it in is labeled Description. The AI reads that text to decide whether to call the tool, so write down when it should be used — describe the situation, not the endpoint. The allowance is per brand: one on the free plan, twenty on a paid plan.
When the Agent Uses a Tool
The agent decides, turn by turn, whether to answer directly or call a tool. In practice:
- Generic factual question ("what are your shipping zones?") → knowledge lookup.
- Personal or account question ("where is my order?") → an API tool that takes the order number.
- Something you scripted ("I'd like to book a slot") → the matching workflow.
- Small talk, or something already in the agent's instructions ("are you a robot?") → a direct answer, no tool.
While a tool runs, the agent keeps listening and talking: it says a very short line, in the language being spoken, that it is still on it. That line is deliberate — it tells the caller that real work is happening.
How long a tool may take is the same as in chat. An API tool waits for its own time limit (5 seconds unless you change it), and a Call an API workflow step waits for the step's own time (8 seconds by default, 30 at most). On a phone call a Call an API step still tries only once.
Answering in the Caller's Language
The brand's material is searched in every language it is written in — so an English knowledge base and a Hungarian one both keep their full search accuracy on the same call.
When the caller speaks a different language from the one your material is written in, the search wording is translated before the lookup so it still matches. It is translated into the brand's own language, set at Brand → Brand settings → General.
When Nothing Matches the Question
Privacy Masking on Calls
Masking is a brand setting: Brand → Brand settings → Advanced → Privacy masking. It hides email addresses, phone numbers and card numbers.
It behaves differently on the phone than in chat, and you should know which. In chat it hides those details before they are sent to the AI provider. On a call the caller's speech has already reached the AI in real time, so what masking protects is what gets stored — the saved transcript and the call's messages in your inbox. It is best-effort, never a guarantee: if the masking service is slow or unavailable the conversation is stored as it was, and nothing records that it happened. It is not encryption. Calls recorded before you switched the toggle on stay as they were.