oneira
OverviewFeaturesCompareSpecsPreorder a Founding Unit · $80
Oneira guides/MCP development

Developer tutorial / MCP servers

Build the server once. Ask from your pocket.

A developer guide to MCP development: how servers expose tools to AI hosts, why one server serves many surfaces, and how Oneira reaches servers you configure.

Preorder a Founding Unit · $80 →See current features →

Product status

What is available now?

  • Working today

    Working today means the expressive face has been demonstrated on a real Oneira device.

  • Connected beta

    Connected beta features require the Oneira app, internet access, a compatible service, and permissions you approve. They can change and may not work with every service.

What you need

Connected beta features need the Oneira app, internet access, a supported service, and permissions you approve. A separate account or fee may also be required.

The short answer

The Model Context Protocol lets you describe a tool once and reach it from many AI applications. Oneira ships MCP support, so a server built for your desktop workflow can also answer a pocket-started request — once you configure and test the connection.

Colorful animated cards representing tools an MCP server can expose to connected AI hosts
Illustrative interface animation / compatibility varies
ProtocolOpen and documented
ServersExpose defined tools
On OneiraMCP support ships
CompatibilityTested, not universal

Oneira Dream Big+

A local AI powerhouse in a star.

Small enough to look playful. Built with a dual AI chip, OLED screen, and 1K camera, plus optional connected experiences you control.

Each $80 Founding Unit includes one Oneira device with a dual AI chip, OLED screen, and 1K camera, plus the selected color and included eye set. Only the final 1K camera module may change before fulfillment. Optional connected services, separate accounts, usage, and third-party fees are not included.

On-device

On-device reactions

Oneira's expressive reactions run on the device, so its face can respond without relying on a cloud request.

See how reactions work ↗
Working prototype

Dual AI chip, OLED screen, and 1K camera

The working prototype includes a dual AI chip, OLED screen, and 1K camera. The dual AI chip, OLED screen, and 1K camera are included in the $80 Founding Unit; only the final 1K camera module may change before fulfillment.

See included hardware ↗
Connected beta

Optional connected services

Optional compatible connected services can extend Oneira when you choose. Setup, permissions, internet access, service terms, accounts, usage, and fees remain separate. Connected experiences work only with compatible services the owner deliberately configures. Oneira does not connect automatically to every service.

See compatible connections ↗
Connected beta

Compatible connected tools

Compatible connected tools are optional and owner-configured. Nothing is preconfigured, and universal compatibility is not promised.

See connected features ↗
Web beta

Dream Big+ photo generation

The 1K camera supports deliberate, owner-triggered experiences. Optional Dream creation can transform a photo when the owner chooses it.

Open Oneira Create ↗
Developer preview

Focused helpers

Optional helper experiences have narrow, visible limits. Broader research, coding, and workspace actions remain in development.

Understand the agent path ↗
The honest boundary

Oneira's focused helpers have narrow, visible limits. Broader research, coding, and approved-workspace actions remain in development.

A simple flow

How it works in three steps.

  1. 01
    01 / Define

    Expose a few narrow, well-named tools

    →
  2. 02
    02 / Prove

    Test read-only calls from a host you already use

    →
  3. 03
    03 / Connect

    Configure the server in the Oneira app with scoped permissions

The protocol

One standard between tools and AI applications.

The Model Context Protocol is an open protocol that standardizes how AI applications connect to tools and data. A server exposes capabilities with defined names and inputs; a host application discovers those capabilities and calls them on the user's behalf.

The documentation and SDKs are public at modelcontextprotocol.io, so nothing on the server side depends on a single vendor. You describe what your service can do once, in one interface, instead of writing a custom integration per assistant.

Why hardware matters

One server, many surfaces, including a pocket.

The return on building a server compounds with every host that can reach it. The same tool that answers in a desktop chat can answer in an editor, an agent workflow, or a physical device, without a new line of server code.

That is the device angle worth designing for. Oneira ships MCP server support, so a request that begins with a deliberate action on a palm-sized device can route through the Oneira app to a server you configured. The server does not need to know the request came from hardware.

Design for it

Small tools read better on a small screen.

A round 466 by 466 pixel display rewards tools that return one usable answer, not a page of raw data. Prefer narrow tools with descriptive names, small structured results, and defaults that make a short request meaningful.

Separate reads from writes at the tool level. A tool that checks a status is safe to call casually; a tool that changes something deserves its own name, its own permission decision, and human approval before it runs from any surface.

On Oneira

Configured connections, applied permissions.

Oneira follows one published pattern: you connect a server, authenticate with your own accounts where the server requires them, and grant specific permissions in the Oneira app. The app routes requests only to servers you pointed it at.

No universal MCP compatibility is promised — transport, authentication, and schema still have to line up, so test each server on a small task before relying on it. The exact configuration screens are not published in this guide; the pattern above is the part that is.

Sources and scope

Official facts, linked.

  • Model Context Protocol — official documentation ↗
  • Oneira MCP servers — capability overview ↗

This independent comparison is not sponsored or endorsed by the compared company. Product names and trademarks belong to their respective owners.

Questions, answered

The details that decide a purchase.

What is MCP development in one sentence?+

It is building or adopting servers that speak the Model Context Protocol, an open standard for connecting AI applications to tools and data, so any compatible host can call your service through one defined interface.

Do I have to build anything Oneira-specific?+

No. You build against the public protocol documentation, and Oneira connects to compatible MCP servers you configure through its app. Compatibility still has to be tested — no universal support is promised.

Which tools should a first server expose?+

Read-only ones. A status check or a lookup proves the whole path with nothing at stake, and write tools can be added later behind their own permissions and human approval.

Will every MCP server work with Oneira?+

No universal compatibility is promised. Transport, authentication, schema, and Oneira app support all have to align, so treat each server as a connection to test before you rely on it.

What does running this cost?+

No Oneira subscription is required for the core product. Server hosting, the services a server talks to, and any accounts behind them are your own, with your own fees.

Keep exploring

Understand the product before you choose it.

Developer guide / physical MCPA physical front door for MCP tools.→Oneira Connect / app + internetYour tools. In your pocket.→Architecture guide / your stack, your choiceWhat is a bring-your-own-AI device?→

Oneira / current availability

Preorder a Founding Unit · $80

500 Drop 01 units remain across 5 currently available colors. Availability is confirmed when secure checkout begins.

Preorder a Founding Unit · $80 →
oneira

Dual AI chip. OLED screen. 1K camera.

ComparePrivacyPreorder termsSupport