coc-reference

Explain the COC framework's five-layer architecture and Human-on-the-Loop workflows.

Updated Apr 2, 2026
One-click install
npx skills add https://github.com/aliciapls/ML-Week-2---Healthcare --skill coc-reference-aliciapls
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: coc-reference
Source: https://github.com/aliciapls/ML-Week-2---Healthcare/tree/main/.claude/skills/28-coc-reference
Command: npx skills add https://github.com/aliciapls/ML-Week-2---Healthcare --skill coc-reference-aliciapls

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a compact reference for understanding the COC framework so AI-assisted development stays disciplined, context-aware, and aligned with Human-on-the-Loop practices.

Core Features & Use Cases

  • Framework overview: Summarizes the five-layer COC architecture and the problem it solves across intent, context, guardrails, instructions, and learning.
  • Concept grounding: Clarifies key ideas such as vibe coding, institutional knowledge, anti-amnesia patterns, and deterministic enforcement.
  • Use case: Use it when you need to explain COC terminology, compare COC to ad hoc coding workflows, or orient an assistant to the framework's operating model.

Quick Start

Ask the assistant to explain the COC framework, its five layers, and how to apply it in AI-assisted software development.

Frequently Asked Questions about coc-reference

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
What is the COC framework for AI-assisted development?

The COC framework is a disciplined operating model for AI-assisted development structured around a five-layer architecture spanning intent, context, guardrails, instructions, and learning to maintain alignment and prevent context loss.

How does Human-on-the-Loop differ from vibe coding?

Human-on-the-loop enforces deterministic guardrails and institutional knowledge engineering, whereas vibe coding relies on ad hoc AI generation without anti-amnesia patterns or structured architectural constraints.

How do I apply anti-amnesia patterns in AI-assisted development?

Apply anti-amnesia patterns by utilizing the COC framework's learning layer to institutionalize context and historical decisions, preventing the AI from losing critical project state during extended software workflows.

What are the five layers of the COC architecture?

The five layers of the COC architecture are intent, context, guardrails, instructions, and learning, which collectively establish disciplined AI-assisted development workflows and deterministic enforcement.

Does the COC framework require specific dependencies to implement guardrails?

No, the COC framework operates without external dependencies, providing a conceptual reference model and mental models to structure guardrails within your existing AI-assisted development environment.

When should I use institutional knowledge engineering instead of ad hoc coding?

Use institutional knowledge engineering instead of ad hoc coding when your AI-assisted development requires strict context-awareness, anti-amnesia patterns, and Human-on-the-Loop oversight to prevent context degradation.