Describe your business. Get a working agent.
No prompt engineering, no config files, no ticket to your IT person. Two or three steps, in your own words — and you see exactly what gets created before anything is created.
The blocker was never the idea
Most owners of a small business already know what their agent should do. They know the five questions customers ask before booking, the two that waste an afternoon, and the answer that has to be exactly right every time.
What stopped them was everything between that knowledge and a working agent: prompts, tools, retrieval settings, a config file, and usually a developer who is busy until next month.
You describe it. We draft it.
Say what your business is in one sentence — the way you would explain it to a new hire, not the way you would write a specification. That is enough for a first draft.
From that sentence the wizard drafts who the agent is, what it does, how it should handle the conversation, and how your documents should be organised for it to search. You read the draft and change anything that is wrong.
The one line we won't write for you
There is exactly one thing the wizard leaves blank on purpose: what the agent must never do.
Never quote a price without a survey. Never promise a delivery date. Never give advice that needs a qualified person. That line depends on how your business carries risk, and nobody outside it can write it honestly — so it is required, and it is yours.
This is also why the boundary is a separate step rather than an optional setting. A limit that is easy to skip is a limit that gets skipped, and it is the field that matters most on the day something goes wrong.
Nothing is created until you have read it
The last step shows the finished agent in full before it exists: the role, the task, the boundary, the knowledge bases it will use. Approve it and it is created; go back and change it as many times as you like first.
You are not committing to it, either. Everything the wizard wrote stays editable — the first version is a starting point, not a decision you are stuck with.
Frequently asked
- Do I need a developer to set up an agent on agent4.io?
- No. The setup wizard in agent4.io asks for a plain-language description of your business and drafts the agent from it — the instructions, the workflow and the knowledge-base structure are written for you. There is no code, no prompt engineering and no configuration file at any step.
- What do I actually have to write myself?
- One sentence describing your business, and the boundary — the line saying what the agent must never do. agent4.io deliberately does not draft the boundary for you, because it is the one decision only the business owner can make, and it is required before the agent can be created.
- What if my trade is not on the template list?
- Describe it in your own words instead. agent4.io drafts a custom agent from that description, and the templates are only a shortcut for common cases — they are not a limit on what the agent can be built for.
- Can I see what will be created before it is created?
- Yes. The final step of the agent4.io wizard shows the whole agent — who it is, what it does, what it must never do, and which knowledge bases it will use — and nothing is created until you confirm it there.
- How long does it take to get a first working agent?
- A few minutes. The agent4.io wizard is two or three steps depending on whether you start from a template, and the result is a working agent you can talk to immediately and refine afterwards.
- Can I change the agent after it has been created?
- Yes. Everything the agent4.io setup wizard writes — the description, the boundary, the knowledge bases and the name — stays editable afterwards, so the first version is a starting point rather than a commitment.
Focused expertise
You don't need another know-it-all chatbot. You need an agent that knows your business cold — and stays inside the boundary you draw.
Per-customer memory
Not one agent for everyone — an agent per customer, holding the whole relationship. Personalization a CRM record can't produce, at a cost that works across your entire book.
Your time back
You can't tell which enquiries deserve your hours until you've already spent an hour finding out. The agent does the finding out — the questions, the eligibility, the paperwork — and what reaches your calendar is a case someone has already briefed.
Agents that act
A chat window that only talks leaves the work to you. This one places the order, books the slot, files the ticket — connected to your systems over MCP and your APIs.
Relationship continuity
In a relationship business, what your team knows about each customer is the asset — and it usually lives in one person's head. Here it accumulates in the company.
Page-aware conversations
Most chat bubbles open on a blank slate and make the visitor explain what they were just reading. This one starts from the page itself, and opens on the question that page actually raises.
Bring your own model
The agent layer and the model layer are separate here. Point agent4.io at any OpenAI-compatible endpoint — a frontier API, a small open model on hardware you control, or several at once — without rebuilding a single agent.