Imperal Docs

Imperal Cloud — the first ICNLI AI Cloud OS

Imperal Cloud is the first ICNLI AI Cloud OS — connect your mail, money, projects, servers and notes under one roof and run them all in plain language.

Your mail is in one app. Your money in another. Your projects, your notes, your calendar, your servers — each behind its own login, its own buttons, its own little wall that says you have to be technical to do this. Dozens of tabs. Dozens of dashboards. One human, pulled in dozens of directions, doing all the connecting by hand.

Imperal Cloud is one roof for all of it — and one language to run it: yours.

You connect the pieces of your life in — mail, money, projects, servers, notes, calendars, anything — and from that moment you stop opening four apps to do one thing. You just say what you want, in your own native language, the way you'd ask a person, and it happens.

"Show me what's overdue, pay the invoice from the design studio, and start the backup before I log off."

One sentence. No new interface to learn, no syntax to memorize, no technical wall — just words, the one interface every human already has.

It works because the intelligence lives inside the system, doing the running-around for you — and held to strict rules the whole time, so it can only ever touch your things the way you allowed. You stay in command; it does the work. This is the first cloud you don't operate — you inhabit it: bring in any context of your life, speak plainly, and let it act.

That inward turn — your life is the context, your words are the interface — is the idea at the very heart of it.

New here? Start with What is Webbee? — plain language, no jargon. Want to see how real this is? The engineering receipts are right below.


The first of its kind

For thirty years, "the cloud" meant servers you rented and dashboards you had to learn. Imperal Cloud is something genuinely new: a cloud with the intelligence living inside it.

The AI runs inside the system — not bolted on top. Every other AI assistant sits outside your tools, asking nicely to use them. Webbee runs inside, so it can actually do the work for you — and it can only ever touch your things the way you allowed.

It grows the way an operating system does. Anyone can teach Webbee a new skill, and you switch on only the ones you want. Your cloud becomes exactly the shape of your life.

It doesn't just wait for you. Webbee can notice when something changes and speak up first — a real assistant, not a prompt box that sits there until you type.

Imperal Cloud is the first ICNLI AI Cloud OS — the first complete implementation of the ICNLI protocol — and we can prove it.

→ For skeptics: Why this is the first ICNLI AI Cloud OS — the six-property test, named-alternative scoring, engineering receipts, falsifiable claim of priority by ICNLI.

→ For builders: What we built that nobody else has — 20 platform-and-agent surfaces with no known precedent.

→ Why Webbee is unique: Cloud-modular + proactive by design — two structural firsts on the ICNLI AI Cloud OS.

💰 Build extensions, earn real money

Imperal Marketplace is a real revenue surface. Every user action through your extension generates credit flow — your developer_share accrues in real time, withdrawable to USD via Stripe Connect.

Pick your starting line

What you can build

Install in one line

pip install imperal-sdk
my_extension/app.py
from imperal_sdk import Extension, ChatExtension, ActionResult
from pydantic import BaseModel, Field

class GreetParams(BaseModel):
    name: str = Field(description="Person to greet")

ext = Extension(
    "hello-world",
    version="1.0.0",
    display_name="Hello World",
    description="Demo extension that greets people by name with a friendly message.",
    icon="icon.svg",
    actions_explicit=True,
)

chat = ChatExtension(ext, "hello_world", description="Hello World — friendly greetings.")

@chat.function("greet", action_type="read", description="Greet someone by name with a friendly message.")
async def greet(ctx, params: GreetParams) -> ActionResult:
    return ActionResult.success(
        data={"text": f"Hello, {params.name}! 🐝"},
        summary=f"Greeted {params.name}",
    )

That's a real, working extension. The LLM will call greet(name="Alex") when a user types "say hi to alex".

Where this docs site lives

You're reading docs.imperal.io — the public reference for everyone building on Imperal Cloud. Sister surfaces: panel.imperal.io (end-user UI), panel.imperal.io/developer (publish + marketplace), auth.imperal.io (OAuth, federation).

Built on hard guarantees

🛡️

Federal-grade

Every action passes through an audit chokepoint. Tenant isolation enforced at the web-kernel layer. Runtime invariants block PRs that weaken them.

🧬

Typed by default

@chat.function uses Pydantic for parameters. The LLM gets a JSON schema, the SDK auto-validates, and a feedback loop retries when args come back wrong.

🌍

Multi-tenant

One extension serves many tenants simultaneously. The web-kernel passes the right user_id everywhere, fail-closed.

Install / upgrade

pip install --upgrade imperal-sdk
Manifest schemav3 (federal contract v4.0)
ValidatorsV14-V24, V31 — most ERROR severity; a few (e.g. V20 effects, the V24 data_model recommendation) are WARN
Anthropic compatibilitySonnet 4.6, Haiku 4.5, Opus 4.8
OpenAI compatibilitygpt-4.1, gpt-5 family, o3, o-series
GitHubimperalcloud/imperal-sdk

For per-release migration notes see the SDK changelog.

All pages on this site

A complete map of everything documented here. Pick anything to jump in.

🚀 Getting Started

🧠 Core Concepts

🛠️ Guides

📐 SDK Reference

📚 Recipes

📖 Reference

🎨 Components showcase

On this page