PlutoAI Agents
Meet the agents that build your app and learn when to use each one
PlutoAI is an agentic platform. You build and maintain your app by talking to agents in plain language: you describe what you want, an agent does the work, and you refine by chatting. This page introduces the agents you can pick and when to use each one.
The build loop
Every agent works the same way underneath. It reasons about your request, makes the change, then checks its own work before calling anything done.
- Reason about the request
The agent works out what you are asking for and plans the change.
- Make the change
It creates and edits the files the change needs and runs any commands it requires.
- Check its own work
It confirms the app still builds, the live preview loads, and every planned step is done.
- Finish or keep going
It will not call a task finished until those checks pass. If more work remains, it continues.
Because the agent does not assume a change worked, a completed change is usually one you can try right away in the preview. If it hits an error it retries sensibly, and if it keeps hitting the same problem it stops and tells you rather than looping forever. You can follow along and watch the live preview update as it works. See Troubleshooting and Testing.
Aris Mini
Aris Mini is the hands-on builder and the default agent. You talk to it directly and it does the work end to end: it plans, writes the code across your project, and checks that the app runs. When it needs a decision from you, it stops and asks. This is the right choice for most projects, from a first landing page to a full app.
Ameca
Ameca is the project lead. It takes a larger request, breaks it into a plan, keeps track of the overall goal and progress, and hands each piece to the right specialist. It reviews that the work is actually done before it finishes. Reach for Ameca on bigger, multi-part builds that you want organized into stages.
The specialists
When Ameca runs, it can call on three focused specialists:
- A1 handles the front-end: pages, components, layout, and design.
- A2 handles the back-end: APIs, data, and databases.
- A3 handles security and deployment.
You do not manage the specialists directly. Ameca assigns them work and combines the results into one coherent change. Because they work in the background, they do not pause to ask questions; when a decision is needed, Ameca is the one that stops and asks. If you want to focus on one area, you can direct a single turn to one specialist from the chat. See Skills.
Which one should I pick
| Situation | Pick |
|---|---|
| A new site or app, most everyday changes | Aris Mini |
| A single focused fix or feature | Aris Mini |
| A large, multi-part build you want organized into stages | Ameca |
| Work that spans front-end, data, and deployment at once | Ameca |
You choose the agent from the agent selector in the chat, and you can switch as a project grows. It is common to start with Aris Mini and move to Ameca as the scope increases. If you are unsure, start with Aris Mini.
You stay in control
At decision points the agent pauses and asks you a real question, with suggested choices or room for your own answer. It only uses a service once you have connected it. You can see the files it writes, and it explains what it changed in the conversation. See Agent permissions.