Agents

An agent is a representative of your company that talks to your customers in the messaging apps they already use. You describe the agent and Anychat handles the conversation: choosing what to do next, writing the reply, and rendering it in whatever rich-messaging features the person's channel supports.

Every agent runs on the same framework, whatever it was built from. Its behavior is meant to be predictable, checkable, and explainable by the person who runs it.

What an agent is made of

An agent is four things, and you supply each one differently.

Part	What it is	Your verb
Behavior	One document: who the agent is, how it sounds, what it never does, what it works toward, and a Playbook for each kind of situation.	Write it.
Knowledge	What the agent knows: Sources, Facts, Things, Media, and business info.	Give it.
Capabilities	What the agent can do beyond talking: book calls, remember people, alert your team, call your systems.	Switch them on.
Channels	Where the agent speaks: RCS, Apple Messages for Business, WhatsApp, SMS.	Connect them.

Behavior and Knowledge divide cleanly. A sentence about tone, sequence, or a limit belongs in Behavior. A fact about your prices, hours, or catalog belongs in Knowledge, where you correct it once and every Playbook gets the correction.

Two more concepts sit alongside the four. Playbooks and Flows give a Playbook a fixed path when the situation calls for one, and Rules are limits the platform enforces whatever the agent would otherwise say.

The smallest working agent is an Identity and one Default Playbook. Sections and capabilities are added as the use case demands.

What the agent reads

The agent only ever reads what you can see.

On every turn the agent works from the Behavior document, what it knows from Knowledge, each enabled capability's short "what this adds" text as printed on its Capabilities card, and a one-line profile of the channel the conversation is on. Nothing is added behind the scenes.

To find out why an agent said something, open the Behavior document, read the capability's text, and look at the same channel line the agent saw. Preview what the agent sees on the Behavior page shows the composed whole for any Playbook and channel, and every real turn keeps a record you can open afterwards. See How a turn works.

A worked example

The concept pages share one example agent. Mia is the sales-development agent for Glow Aesthetics, a med spa. Mia answers questions about treatments, qualifies people who are considering a visit, books consultation calls with a member of the Glow team, and gives people the clinic's contact details when they ask for a human. Mia talks on RCS and falls back to SMS.

Mia's Behavior has five Playbooks: Welcome (the default), Answer treatment questions, Qualify, Book a consultation call, and Talk to a person. Three capabilities are switched on: Scheduling, People, and Notifications. Knowledge holds the treatment pages from glowaesthetics.com and a catalog of eight treatments.

One turn, in words

A person on RCS writes to Mia: "Do you do laser hair removal, and could I get a call this week?"

1. Anychat checks the message first. This person has not opted out, and the channel allows a reply right now.
2. Anychat picks the Playbook that fits. The message is about a treatment and about booking. Answer treatment questions matches first, and its when-to-use line says to hand over to Book a consultation call once the person is ready.
3. Anychat composes what the agent reads this turn: the Behavior document with the active Playbook's full body and only the names and when-to-use lines of the others, the Knowledge and Scheduling "what this adds" text, the RCS channel line, this person's record, and the conversation so far.
4. The agent looks up laser hair removal in Glow's Sources, then writes its reply: a short answer and a question about whether the person would like to see available times.
5. Anychat checks the reply against the Rules, shapes it for RCS as a text bubble and two suggestion chips, and sends it.
6. The turn is recorded: what was read, what was looked up, what was said, and what was sent.

How a turn works takes each of these apart.

What's next

• Behavior: the document, section by section.
• Create an agent: start from a template and have a working agent in the emulator.
• Agent templates: the complete starting points Anychat curates.
• Agent API: everything above, programmatically.