bkit 面向 OpenAI Codex 的 Vibecoding Kit - PDCA + 上下文工程用于 AI 原生开发
概览
bkit-codex 是一个 OpenAI Codex CLI 扩展，能够改变你用 AI 构建软件的方式。它通过 PDCA（计划-做-检查-行动）方法论和上下文工程架构，提供结构化的开发流程、自动文档和智能代码协助。 是系统地策划上下文标记，以实现最佳的大型语言模型推断——超越简单的提示词制作，构建完整系统，持续指导人工智能行为。 ，提供跨多个AI编码平台的系统化上下文管理系统。
概述（AI 翻译）
PDCA 方法论 + 面向 AI 原生开发的上下文工程
bkit-codex 是一个 OpenAI Codex CLI 扩展，能够改变你用 AI 构建软件的方式。它通过 PDCA（计划-做-检查-行动）方法论和上下文工程架构，提供结构化的开发流程、自动文档和智能代码协助。
什么是情境工程？
上下文工程是系统地策划上下文标记以实现最佳大型语言模型推理——超越简单的提示编写，构建持续指导AI行为的完整系统。
**bkit 是上下文工程的实用实现，提供跨多个 AI 编码平台的系统化上下文管理系统。
bkit-codex 的上下文工程架构
bkit-codex 通过三个相互关联的层实现上下文工程：
| 层 | 组件 | 数量 | 目的 |
|---|---|---|---|
| 领域知识 | 技能 | 27 | 通过渐进披露按需激活的结构化专家知识 |
| 行为规则AGENTS.md | 2个文件 | 全局+项目规则，必须/始终/绝不执行 |
本页由 AI 翻译，技术细节请以原文为准。
README
以下为原文 README 节选，完整内容请查看原文。
bkit-codex - Vibecoding Kit (OpenAI Codex Edition)
PDCA methodology + Context Engineering for AI-native development
bkit-codex is an OpenAI Codex CLI extension that transforms how you build software with AI. It provides structured development workflows, automatic documentation, and intelligent code assistance through the PDCA (Plan-Do-Check-Act) methodology and Context Engineering architecture.
What is Context Engineering?
Context Engineering is the systematic curation of context tokens for optimal LLM inference – going beyond simple prompt crafting to build entire systems that consistently guide AI behavior.
Traditional Prompt Engineering:
"The art of writing good prompts"
Context Engineering:
"The art of designing systems that integrate prompts, tools, and state
to provide LLMs with optimal context for inference"
bkit is a practical implementation of Context Engineering, providing a systematic context management system across multiple AI coding platforms.
bkit-codex’s Context Engineering Architecture
bkit-codex implements Context Engineering through three interconnected layers:
| Layer | Components | Count | Purpose |
|---|---|---|---|
| Domain Knowledge | Skills | 27 | Structured expert knowledge activated on-demand via progressive disclosure |
| Behavioral Rules | AGENTS.md | 2 files | Global + Project rules with MUST/ALWAYS/NEVER enforcement |
| State Management | MCP Tools + Lib Modules | 16 + 80 fn | PDCA status tracking, intent detection, template management, memory persistence, context recovery |
Architecture
3-Tier Context Strategy
+------------------------------------------------------------------+
| Tier 1: AGENTS.md |
| (Always Loaded) |
| |
| Global AGENTS.md (~3.8 KB) Project AGENTS.md (~2.0 KB) |
| - Session initialization - Level-specific guidance |
| - 3 Core Principles - Key skills reference |
| - PDCA workflow rules - Response format rules |
| - MCP tools reference - PDCA status location |
| - Code quality standards |
| |
| Total: ~5.8 KB / 32 KB limit (18% used, 82% available) |推荐工具
换一个关键词，或者移除筛选条件。
安装
npx skillfish add ww-w-ai/bkit-codex