🐝 Imperal Quantum SDK (imperal-sdk)
The Official Foundation AI Cloud OS Framework for Imperal Cloud & Webbee 🐝
Write autonomous, context-aware AI tools and extensions in pure, idiomatic Python. Zero manifests. 0 lines of JSON boilerplate. Automatic reactive UI synthesis. Ambient cloud context. Multi-surface liquid morphing across Terminal, Web Panel, Telegram, and Voice. Python 3.6 to 3.14+ universal support.
Documentation · ICNLI Protocol Spec · Imperal Cloud Panel · Marketplace
🌌 The Quantum Leap: Forget Legacy AI Agent Tooling
In early AI agent frameworks and Web2 SDKs, building tools for agents degenerated into bureaucratic overhead:
- ❌ Manifest Hell: Manually maintaining hundreds of lines of fragile
manifest.jsonfiles. - ❌ Redundant Schema Declarations: Copy-pasting parameters between Python signatures and JSON Schema definitions.
- ❌ Frontend Tax: Needing dedicated UI engineers just to render an agent's structured response in a web dashboard.
- ❌ Surface Fragmentation: What worked in the terminal broke in Telegram and looked horrible on the web.
- ❌ Prompt Token Waste: Leaking internal runtime parameters (
context,user_id,tokens) directly into LLM function-calling schemas.
⚡ Enter ICNLI Quantum SDK v6.0
Imperal Quantum SDK delivers to autonomous cloud agents what FastAPI brought to modern web APIs — elevating cloud application development into an ultra-clean, quantum plane:
- 🚀 Zero-Manifest (0 Lines of JSON): Manifest definitions, parameter schemas, docstrings for LLM planning, RBAC security scopes, and billing tiers are extracted dynamically from standard Python function signatures and type hints.
- 🎨 Auto-IR (UI-as-Data): Simply return standard Python dictionaries, dataclasses, or lists. The Imperal Cloud runtime automatically projects them into reactive Declarative UI components (cards, metric grids, data tables) without writing a single line of CSS or frontend code.
- 🔮 Ambient Context Injection: Seamlessly access session identity (
Context), encrypted key-value storage (Store), persistent object storage (Storage), and the native AI engine (AI) without polluting the LLM's function calling schema. - 🔀 Liquid Multi-Surface Morphing: One codebase transparently adapts its presentation across Terminal TUI (Webbee Code), Web Console (Imperal Panel), Telegram Messenger, and Ambient Voice.
- 🛡️ Universal Python Compatibility (3.6 – 3.14+): From legacy MCP environments running Python 3.6 up to cutting-edge Python 3.14 runtimes, the core SDK operates with 100% backward compatibility and zero overhead.
- 💎 100% Backward Compatible: Full support for classic enterprise extensions built on
imperal_sdk.Extension(validated by 1,620+ automated tests).
🚀 Quickstart in 60 Seconds
1. Installation
pip install imperal-sdk
2. Your First Cloud App in 15 Lines (server_pulse.py)
from imperal_sdk import App, Context
# Instantiate your Quantum App — your manifest is already generated!
app = App("server-pulse", name="Server Pulse", category="devops")
@app.tool(pricing=5, destructive=False)
def check_host(host: str, port: int = 443, ctx: Context = None) -> dict:
"""Check server availability and SSL certificate expiration in real time."""
# Pure Python business logic:
is_online = True
ssl_days = 89
# Return pure data — Imperal Cloud synthesizes the reactive UI on the fly!
return {
"status": "online" if is_online else "down",
"host": host,
"ssl_days": ssl_days,
"metric": f"{ssl_days} days left"
}
What happened under the hood?
- ✅ Valid ICNLI Manifest v6.0: Automatically synthesized with zero JSON.
- ✅ Type-to-Schema Translation:
hostandportbecame typed JSON Schema properties with defaults. - ✅ Docstring Extraction: The docstring was converted into the LLM classifier and planner description.
- ✅ Security Scopes: Granular RBAC scope
server-pulse:check_hostwas registered automatically. - ✅ Ambient Context:
ctxwas hidden from the LLM prompt while remaining injected at execution time. - ✅ Auto-IR Projection: The returned dictionary automatically renders in
panel.imperal.ioas a styled Metric Card with badges and metrics!
💎 Core Quantum Superpowers
1. ⚡ Automatic Schema Synthesis (Type-to-Schema Engine)
Write clean, idiomatic Python with standard typing. The SDK extracts complete JSON Schema definitions:
from typing import List, Optional
@app.tool()
async def deploy_services(
environment: str,
replicas: int = 3,
tags: Optional[List[str]] = None,
dry_run: bool = False
) -> dict:
"""Deploy microservice instances to the specified cluster zone with autoscaling."""
...
Synthesized parameter schema:
{
"name": "deploy_services",
"description": "Deploy microservice instances to the specified cluster zone with autoscaling.",
"parameters": {
"type": "object",
"properties": {
"environment": { "type": "string" },
"replicas": { "type": "integer", "default": 3 },
"tags": { "type": "array", "items": { "type": "string" } },
"dry_run": { "type": "boolean", "default": false }
},
"required": ["environment"]
}
}
2. 🎨 Auto-IR: UI-as-Data (Frontend without Frontend Code)
The Imperal Cloud Kernel inspects your return payloads and projects them into Declarative UI:
| Python Return Value | Rendered Appearance in Imperal Panel |
|---|---|
{"status": "ok", "metric": "99.98%", ...} |
🎴 Metric Card: Interactive card with status badge, headline metric, and details |
[{"id": 1, "name": "node-a", "cpu": 14}, ...] |
📊 Data Table: Sortable, responsive table with auto-detected columns |
{"_ui": {"type": "custom", ...}} |
🧩 Declarative IR: Full customization using Imperal UI Kit primitives |
3. 🔮 Ambient Context: Invisible Access to the Entire Cloud OS
The ctx: Context argument is hidden from the LLM schema (saving prompt tokens), but grants full platform access at runtime:
@app.tool()
async def analyze_anomalies(cluster_id: str, ctx: Context = None) -> dict:
# 1. Access high-speed key-value store (Redis):
last_run = await ctx.store.get(f"scan:{cluster_id}")
# 2. Query the native AI brain (LLM cascade):
verdict = await ctx.ai.complete(f"Diagnose cluster status: {cluster_id}")
# 3. Read authenticated identity and enterprise tenant boundaries:
actor_email = ctx.user.email
tenant_id = ctx.user.tenant_id
return {
"status": "analyzed",
"verdict": verdict.text,
"initiated_by": actor_email,
"metric": "Clean"
}
4. 🔀 Multi-Surface Liquid Morphing
Webbee operates natively across all surfaces. The same tool presentation is automatically morphed:
- Terminal (Webbee Code TUI): High-density formatted output, ASCII tables, and numbered hotkey shortcuts (
[1] Execute,[2] Abort). - Web Console (Imperal Panel): Interactive Declarative UI cards, modal dialogs, and real-time streaming widgets.
- Telegram Messenger: Concise mobile cards with inline button callbacks.
- Ambient Voice: Spoken concise executive summaries.
5. 🔗 Cross-Extension Unix Piping
Tools across completely different extensions can be piped together through plain language:
"Webbee, fetch domains from DNS checker, pass them to SSL auditor, and ping me on Telegram if expiring soon."
Every tool declaring typed inputs and outputs participates in deterministic, automated pipeline chaining.
🧪 Hermetic Local Testing (Zero Network Required)
Imperal SDK includes a built-in autonomous mock suite (imperal_sdk.testing.autonomous_mock):
import pytest
from imperal_sdk.testing.autonomous_mock import MockContext
from server_pulse import check_host
def test_check_host_locally():
mock_ctx = MockContext(user_id="imp_u_test", role="admin")
result = check_host(host="imperal.io", port=443, ctx=mock_ctx)
assert result["status"] == "online"
assert "_ui" in result
assert result["_ui"]["type"] == "metric_card"
assert result["_ui"]["badge"] == "online"
Run test suite:
pytest -v
🏗️ Imperal Cloud OS Architecture
┌───────────────────────────────┐
│ Webbee Brain │
│ (Agentic AI Reasoning) │
└───────────────┬───────────────┘
│ Intent & Tool Calls
▼
┌───────────────────────────────┐
│ ICNLI Quantum SDK v6.0 │
│ imperal_sdk.App / Extension │
└───────────────┬───────────────┘
│ Declarative IR & Telemetry
┌────────────────────┼────────────────────┐
▼ ▼ ▼
┌──────────────┐ ┌──────────────┐ ┌──────────────┐
│ Imperal Panel│ │ Webbee Code │ │ Telegram │
│ (Web UI) │ │ (Terminal) │ │ (Messenger) │
└──────────────┘ └──────────────┘ └──────────────┘
📜 Compatibility & Specifications
- Python Runtime: Universal support for Python 3.6, 3.7, 3.8, 3.9, 3.10, 3.11, 3.12, 3.13, 3.14+.
- Footprint: Ultra-lightweight core with lazy attribute resolution (PEP 562 with Python 3.6 fallback). Sub-15ms cold start.
- Standards: 100% compliant with the ICNLI Open Protocol (CC BY-SA 4.0).
- License: Apache-2.0.
Imperal Cloud — The World's First Decentralized ICNLI AI Cloud OS.
Crafted with 💛 and 🐝 by Imperal, Valentin Scerbacov, and the open-source community.
1 maintainer
Metadata
1 maintainer
Release files for imperal-sdk 6.2.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| imperal_sdk-6.2.0.tar.gz | 890.2 kB | With provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| imperal_sdk-6.2.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 1.3 MB
Release files / imperal_sdk-6.2.0.tar.gz
| Download URL | imperal_sdk-6.2.0.tar.gz |
|---|---|
| Size | 890.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
dff06d4094095c0141e22b859e79efa352c57d8e7f20825b5efcf8c3c9f46a90
|
|
BLAKE2b-256 checksum How to use checksums |
f9a6b4d1bf2f6c94c4817ca983d856842febef2bf2b527038e5e69ae7b88ad04
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 20, 2026.
Transparency logRelease files / imperal_sdk-6.2.0-py3-none-any.whl
| Download URL | imperal_sdk-6.2.0-py3-none-any.whl |
|---|---|
| Size | 415.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
6d2bea77062b4702e69758087250f8c00f4df8f95274d24211d598d6eb1aa063
|
|
BLAKE2b-256 checksum How to use checksums |
fe01f29fca97b700094aa72af3fbdd1f035e90566b9af7e60c4e17a0755dce50
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 20, 2026.
Transparency logSecurity for imperal-sdk 6.2.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
Full provenance
All 2 files in this release have been uploaded with provenance metadata.
Attestation for imperal_sdk-6.2.0.tar.gz
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 20, 2026.
Transparency logA note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
