apcore

AI-Perceivable module standard

About

apcore is an AI-Perceivable module standard that makes every interface naturally perceivable and understandable by AI through enforced Schema definitions and behavioral annotations. It ensures every module is inherently understandable by AI through enforced schemas, behavioral annotations, and rich intent metadata. The framework features a secured execution lifecycle with a unified APCore client, built-in Event System, and pattern-based ACL. It includes a phase-based Approval System for human-in-the-loop enforcement and a streaming execution protocol for real-time interactions. Built with enterprise-grade observability (ErrorHistory, UsageCollector) and system modules, apcore enables seamless cross-language consistency between Python, TypeScript, and Rust.

Features

Unified APCore Client: Single entry point for registry, executor, events, and control

Secured Execution Lifecycle: Robust execution with preflight validation and dual-timeout

Built-in Event System: Framework-wide EventEmitter and EventSubscriber support

System Modules: Integrated health, usage, manifest, and runtime control modules

Schema-Driven: Mandatory input_schema, output_schema, and behavioral annotations

Approval System: Runtime human-in-the-loop enforcement for sensitive operations

Streaming Protocol: Standardized incremental output for real-time interactions

AI Intent Metadata: Semantic hints like x-when-to-use and x-workflow-hints

Observability: Built-in ErrorHistory (ring buffer) and UsageCollector tracking

Cross-Language SDKs: Identical API and behavior for Python, TypeScript, and Rust

Get Started

Python Implementation

Python reference implementation of the apcore specification.

View Repository PyPI Package

Install

pip install apcore

Quick Start

from apcore import APCore

client = APCore()

@client.module(id="math.add", description="Add two numbers")
def add(a: int, b: int) -> dict:
    return {"sum": a + b}

result = client.call("math.add", {"a": 10, "b": 5})