coc-reference

Clarify the COC framework for disciplined AI-assisted development workflows.

Updated Apr 2, 2026
One-click install
npx skills add https://github.com/myeohVmzrka92/disease-risk-controller-S2 --skill coc-reference-myeohvmzrka92
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: coc-reference
Source: https://github.com/myeohVmzrka92/disease-risk-controller-S2/tree/main/.claude/skills/28-coc-reference
Command: npx skills add https://github.com/myeohVmzrka92/disease-risk-controller-S2 --skill coc-reference-myeohvmzrka92

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The COC Framework Reference explains how to replace vibe coding with disciplined, institutionally aware AI-assisted development by preserving context, enforcing guardrails, and reducing amnesia, convention drift, and security blindness.

Core Features & Use Cases

  • Five-Layer Architecture: Summarizes intent, context, guardrails, instructions, and learning as a complete operating model for Claude Code.
  • Development Methodology Reference: Helps teams reason about Human-on-the-Loop workflows, framework-first implementation, and approval-gated execution.
  • Use Case: A software engineer can use this skill to understand how the COC system structures agents, skills, rules, commands, and learning loops before making architecture or implementation decisions.

Quick Start

Use the coc-reference skill to explain the COC framework and apply its five-layer architecture to my current AI-assisted development task.

Frequently Asked Questions about coc-reference

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

FAQPage Schema
How do I stop vibe coding and prevent context amnesia in AI-assisted development?

To stop vibe coding and prevent context amnesia, apply the COC framework's five-layer architecture to preserve institutional knowledge, enforce guardrails, and maintain consistent context across AI-assisted development sessions.

What is the Human-on-the-Loop workflow for disciplined AI code generation?

The Human-on-the-Loop workflow is an approval-gated execution model within the COC framework that structures AI code generation by enforcing guardrails and human oversight, replacing autonomous AI coding with disciplined, institutionally aware development.

How do I structure agents, skills, and instructions for Claude Code projects?

Structure agents, skills, and instructions for Claude Code by applying the COC framework's five-layer architecture, which organizes intent, context, guardrails, instructions, and learning into a complete operating model for code-generation systems.

Does the COC framework support convention drift and security blindness prevention?

Yes, the COC framework supports convention drift and security blindness prevention by applying structured guardrails and learning loops, ensuring AI-assisted development remains disciplined and institutionally aware.

Can I use context engineering patterns for existing AI development tasks?

Yes, you can apply context engineering patterns to existing tasks by using the COC framework reference to map your current AI development workflow onto its five-layer architecture for consistent code-generation system reasoning.