{"id":45450040,"url":"https://github.com/popup-studio-ai/bkit-codex","last_synced_at":"2026-02-22T05:01:19.685Z","repository":{"id":338468678,"uuid":"1157969371","full_name":"popup-studio-ai/bkit-codex","owner":"popup-studio-ai","description":"bkit Vibecoding Kit for OpenAI Codex - PDCA + Context Engineering for AI-native development","archived":false,"fork":false,"pushed_at":"2026-02-14T17:04:33.000Z","size":317,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-02-15T00:56:32.465Z","etag":null,"topics":["ai-agents","ai-driven-work","ai-native-development","bkit","codex","multi-language","pdca","pipeline","plugin-repo"],"latest_commit_sha":null,"homepage":"https://bkit.ai","language":"JavaScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/popup-studio-ai.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2026-02-14T15:38:11.000Z","updated_at":"2026-02-14T17:03:59.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/popup-studio-ai/bkit-codex","commit_stats":null,"previous_names":["popup-studio-ai/bkit-codex"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/popup-studio-ai/bkit-codex","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/popup-studio-ai%2Fbkit-codex","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/popup-studio-ai%2Fbkit-codex/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/popup-studio-ai%2Fbkit-codex/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/popup-studio-ai%2Fbkit-codex/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/popup-studio-ai","download_url":"https://codeload.github.com/popup-studio-ai/bkit-codex/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/popup-studio-ai%2Fbkit-codex/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29705523,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-22T03:17:42.375Z","status":"ssl_error","status_checked_at":"2026-02-22T03:17:31.622Z","response_time":110,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-agents","ai-driven-work","ai-native-development","bkit","codex","multi-language","pdca","pipeline","plugin-repo"],"created_at":"2026-02-22T05:01:18.413Z","updated_at":"2026-02-22T05:01:19.672Z","avatar_url":"https://github.com/popup-studio-ai.png","language":"JavaScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"# bkit-codex - Vibecoding Kit (OpenAI Codex Edition)\n\n[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)\n[![Codex CLI](https://img.shields.io/badge/Codex%20CLI-v0.100.0+-blue.svg)](https://github.com/openai/codex)\n[![Version](https://img.shields.io/badge/Version-1.0.0-green.svg)](CHANGELOG.md)\n[![Author](https://img.shields.io/badge/Author-POPUP%20STUDIO-orange.svg)](https://popupstudio.ai)\n\n\u003e **PDCA methodology + Context Engineering for AI-native development**\n\nbkit-codex is an [OpenAI Codex CLI](https://github.com/openai/codex) 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.\n\n---\n\n## What is Context Engineering?\n\n**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.\n\n```\nTraditional Prompt Engineering:\n  \"The art of writing good prompts\"\n\nContext Engineering:\n  \"The art of designing systems that integrate prompts, tools, and state\n   to provide LLMs with optimal context for inference\"\n```\n\n**bkit is a practical implementation of Context Engineering**, providing a systematic context management system across multiple AI coding platforms.\n\n### bkit-codex's Context Engineering Architecture\n\nbkit-codex implements Context Engineering through three interconnected layers:\n\n| Layer | Components | Count | Purpose |\n|-------|-----------|-------|---------|\n| **Domain Knowledge** | Skills | 27 | Structured expert knowledge activated on-demand via progressive disclosure |\n| **Behavioral Rules** | AGENTS.md | 2 files | Global + Project rules with MUST/ALWAYS/NEVER enforcement |\n| **State Management** | MCP Tools + Lib Modules | 16 + 80 fn | PDCA status tracking, intent detection, template management, memory persistence, context recovery |\n\n---\n\n## Architecture\n\n### 3-Tier Context Strategy\n\n```\n+------------------------------------------------------------------+\n|                    Tier 1: AGENTS.md                              |\n|                    (Always Loaded)                                |\n|                                                                  |\n|  Global AGENTS.md (~3.8 KB)    Project AGENTS.md (~2.0 KB)      |\n|  - Session initialization      - Level-specific guidance         |\n|  - 3 Core Principles           - Key skills reference            |\n|  - PDCA workflow rules          - Response format rules           |\n|  - MCP tools reference          - PDCA status location            |\n|  - Code quality standards                                        |\n|                                                                  |\n|  Total: ~5.8 KB / 32 KB limit (18% used, 82% available)         |\n+------------------------------------------------------------------+\n                              |\n                              v\n+------------------------------------------------------------------+\n|                    Tier 2: SKILL.md                               |\n|                    (On-Demand Loading)                            |\n|                                                                  |\n|  Phase 1: name + description only (metadata scan)                |\n|  Phase 2: Full SKILL.md body (when skill is activated)           |\n|  Progressive Disclosure minimizes context usage                  |\n+------------------------------------------------------------------+\n                              |\n                              v\n+------------------------------------------------------------------+\n|                    Tier 3: references/                            |\n|                    (Deep Dive on Request)                         |\n|                                                                  |\n|  ~30 reference files (templates, patterns, checklists)           |\n|  Loaded only when AI explicitly requests a reference file        |\n+------------------------------------------------------------------+\n```\n\n### Architecture Paradigm Shift\n\n```\nbkit-claude-code (Hook-Driven, 100% Auto)\n  Hook Event --\u003e Script (stdin/stdout) --\u003e decision: allow/block\n  = System controls the AI\n\nbkit-codex (Instruction-Driven, ~70% Auto)\n  AGENTS.md Rule --\u003e AI reads --\u003e AI calls MCP Tool --\u003e MCP returns guidance\n  = AI voluntarily follows rules\n```\n\n### Component Map\n\n```\nbkit-codex/\n|-- .agents/skills/          # 27 Codex Agent Skills\n|   |-- bkit-rules/          # Core rules reference\n|   |-- pdca/                # PDCA workflow management\n|   |-- bkit-templates/      # Template selection\n|   |-- starter/             # Beginner guidance\n|   |-- dynamic/             # Fullstack guidance\n|   |-- enterprise/          # Enterprise guidance\n|   |-- development-pipeline/# 9-phase overview\n|   |-- phase-1-schema/ ... phase-9-deployment/\n|   |-- code-review/         # Code quality analysis\n|   |-- zero-script-qa/      # Log-based testing\n|   |-- mobile-app/          # React Native, Flutter, Expo\n|   |-- desktop-app/         # Electron, Tauri\n|   |-- codex-learning/      # Codex CLI guide\n|   |-- plan-plus/           # Brainstorming-enhanced planning\n|   +-- bkend-*/             # bkend.ai ecosystem (5 skills)\n|\n|-- packages/mcp-server/     # MCP Server (zero external dependencies)\n|   |-- index.js             # STDIO transport (JSON-RPC 2.0)\n|   |-- src/server.js        # Request dispatcher\n|   |-- src/tools/           # 16 MCP tool implementations\n|   +-- src/lib/             # Core library (~80 functions)\n|       |-- core/            # config, cache, file, path\n|       |-- pdca/            # status, level, phase, automation, template\n|       |-- intent/          # language, trigger, ambiguity\n|       +-- task/            # classification, creator\n|\n|-- AGENTS.md                # Project AGENTS.md (sample)\n|-- agents.global.md         # Global AGENTS.md (install to ~/.codex/)\n|-- bkit.config.json         # Centralized configuration\n|-- install.sh               # Unix/Mac installer\n|-- install.ps1              # Windows installer\n|-- scripts/sync-deploy.sh   # Dev-to-deploy sync utility\n|-- docs/                    # Documentation\n+-- .github/workflows/       # CI/CD (npm publish)\n```\n\n---\n\n## Features\n\n- **PDCA Methodology** -- Structured development with automatic documentation and phase enforcement\n- **16 MCP Tools** -- Session management, intent detection, PDCA workflow, template generation, memory persistence\n- **27 Agent Skills** -- Domain-specific knowledge with progressive disclosure to save context tokens\n- **9-Stage Development Pipeline** -- From schema design to deployment\n- **3 Project Levels** -- Starter (static), Dynamic (fullstack), Enterprise (microservices)\n- **8-Language Support** -- EN, KO, JA, ZH, ES, FR, DE, IT with auto-detection\n- **Zero Dependencies** -- MCP server built with pure Node.js (no `node_modules`)\n- **Context Recovery** -- Automatic state reconstruction after context compaction (recovery mode)\n- **Task Chain** -- Linked PDCA task progression with automatic phase-to-phase advancement\n- **Plan Plus** -- Brainstorming-enhanced planning with intent discovery and YAGNI review\n- **Level-Specific Templates** -- Enterprise design template with 11 sections (security, observability, cost analysis)\n- **Ambiguity Detection** -- Scores user prompts 0-100 and generates clarifying questions when needed\n- **Session Memory** -- Persistent context across sessions via `docs/.bkit-memory.json`\n- **Evaluator-Optimizer Pattern** -- Automatic gap analysis and fix cycles (max 5 iterations, 90% threshold)\n- **Cross-Platform Install** -- One-line installation for Unix/Mac and Windows\n\n---\n\n## Quick Start\n\n\u003e **Note**: bkit-codex is designed for **OpenAI Codex CLI**. For Claude Code, see [bkit-claude-code](https://github.com/popup-studio-ai/bkit-claude-code). For Gemini CLI, see [bkit-gemini](https://github.com/popup-studio-ai/bkit-gemini).\n\n### Prerequisites\n\n- [OpenAI Codex CLI](https://github.com/openai/codex) v0.100.0 or later\n- [Node.js](https://nodejs.org/) v20+\n- [Git](https://git-scm.com/)\n\n### Installation\n\n#### Unix / macOS (Recommended)\n\n```bash\ncurl -fsSL https://raw.githubusercontent.com/popup-studio-ai/bkit-codex/main/install.sh | bash\n```\n\n#### Windows (PowerShell)\n\n```powershell\nirm https://raw.githubusercontent.com/popup-studio-ai/bkit-codex/main/install.ps1 | iex\n```\n\n#### Manual Installation\n\n```bash\n# 1. Clone into project\ngit clone --depth 1 https://github.com/popup-studio-ai/bkit-codex.git .bkit-codex\n\n# 2. Link skills\nmkdir -p .agents/skills\nfor skill in .bkit-codex/.agents/skills/*/; do\n  ln -sf \"../../.bkit-codex/.agents/skills/$(basename $skill)\" \".agents/skills/$(basename $skill)\"\ndone\n\n# 3. Copy AGENTS.md\ncp .bkit-codex/AGENTS.md ./AGENTS.md\n\n# 4. Configure MCP server\nmkdir -p .codex\ncat \u003e .codex/config.toml \u003c\u003c 'EOF'\n[mcp_servers.bkit]\ncommand = \"node\"\nargs = [\"./.bkit-codex/packages/mcp-server/index.js\"]\nstartup_timeout_sec = 10\ntool_timeout_sec = 60\nrequired = true\nEOF\n\n# 5. Create PDCA directories\nmkdir -p docs/01-plan/features docs/02-design/features docs/03-analysis docs/04-report\n```\n\n### Verify Installation\n\n```bash\n# Check skills are linked\nls .agents/skills/\n\n# Check MCP config\ncat .codex/config.toml\n\n# Start Codex and invoke bkit\n$pdca status\n```\n\n---\n\n## Usage\n\n### PDCA Workflow\n\n```bash\n$pdca plan \u003cfeature\u003e     # Create plan document\n$pdca design \u003cfeature\u003e   # Create design document\n$pdca do \u003cfeature\u003e       # Implementation guide\n$pdca analyze \u003cfeature\u003e  # Run gap analysis\n$pdca iterate \u003cfeature\u003e  # Auto-fix with Evaluator-Optimizer pattern\n$pdca report \u003cfeature\u003e   # Generate completion report\n$pdca status             # Check current PDCA status\n$pdca next               # Guide to next PDCA step\n```\n\n### Project Initialization\n\n```bash\n$starter      # Static website (Starter level)\n$dynamic      # Fullstack with BaaS (Dynamic level)\n$enterprise   # Microservices with K8s (Enterprise level)\n```\n\n### Development Pipeline\n\n```bash\n$development-pipeline    # 9-phase pipeline overview\n$phase-1-schema          # Schema \u0026 terminology definition\n$phase-2-convention      # Coding conventions\n$phase-3-mockup          # UI/UX mockups\n$phase-4-api             # API design\n$phase-5-design-system   # Design system\n$phase-6-ui-integration  # UI integration\n$phase-7-seo-security    # SEO \u0026 security\n$phase-8-review          # Code review\n$phase-9-deployment      # Deployment\n```\n\n### Quality \u0026 Learning\n\n```bash\n$code-review             # Code quality analysis\n$zero-script-qa          # Log-based testing (Zero Script QA)\n$codex-learning          # Codex CLI configuration guide\n```\n\n---\n\n## PDCA Workflow\n\nThe core development cycle that bkit enforces:\n\n```\nPlan ──\u003e Design ──\u003e Do ──\u003e Check ──\u003e Act ──\u003e Report\n  │                          │         │\n  │    (design required)     │    (if \u003c 90%)\n  │                          │         │\n  X──── skip to Do ────X     +─────────+\n                              iteration loop\n                              (max 5 cycles)\n```\n\n| Phase | Action | Deliverable |\n|-------|--------|-------------|\n| **Plan** | Define goals, scope, success criteria | `docs/01-plan/features/{feature}.plan.md` |\n| **Design** | Architecture, data model, API spec | `docs/02-design/features/{feature}.design.md` |\n| **Do** | Implementation with pre/post-write checks | Source code matching design |\n| **Check** | Gap analysis (design vs. implementation) | `docs/03-analysis/{feature}.analysis.md` |\n| **Act** | Auto-fix gaps until match rate \u003e= 90% | Updated source code |\n| **Report** | Completion report with metrics | `docs/04-report/{feature}.report.md` |\n\n### Key Rules\n\n- **Design is mandatory**: You cannot skip from Plan directly to Do\n- **Pre-write check**: MCP tool validates design document existence before code changes\n- **Post-write guidance**: MCP tool suggests gap analysis after significant changes (\u003e10 new lines or \u003e20 modified lines)\n- **Iteration loop**: Check and Act repeat until match rate reaches 90% (max 5 iterations)\n- **Context recovery**: After context compaction, call `bkit_get_status` with `mode: \"recovery\"` to reconstruct state\n- **Task chain**: Each feature creates a linked Plan-\u003eDesign-\u003eDo-\u003eCheck-\u003eReport task chain with automatic advancement\n\n---\n\n## Project Levels\n\nAutomatic detection based on project directory structure:\n\n| Level | Description | Stack | Auto-Detection | Pipeline Phases |\n|-------|-------------|-------|----------------|:---------------:|\n| **Starter** | Static websites, portfolios | HTML, CSS, JS | Default (no special files) | 1 → 2 → 3 → 6 → 9 |\n| **Dynamic** | Fullstack applications | Next.js, BaaS | `docker-compose.yml`, `.mcp.json`, `api/` | 1 → 2 → 3 → 4 → 5 → 6 → 7 → 9 |\n| **Enterprise** | Microservices architecture | K8s, Terraform, MSA | `kubernetes/`, `terraform/` directories | All 9 phases |\n\n---\n\n## MCP Tools (16)\n\nThe MCP server provides 16 tools via JSON-RPC 2.0 over STDIO with **zero external dependencies**:\n\n| Tool | Category | Purpose |\n|------|----------|---------|\n| `bkit_init` | Session | Initialize session, detect level, load PDCA status, compact summary |\n| `bkit_analyze_prompt` | Intent | Detect language, match triggers, score ambiguity (8 languages) |\n| `bkit_get_status` | PDCA | Retrieve current PDCA status with recommendations (supports recovery mode) |\n| `bkit_pre_write_check` | PDCA | Pre-write compliance check (design document existence) |\n| `bkit_post_write` | PDCA | Post-write guidance (gap analysis suggestions) |\n| `bkit_complete_phase` | PDCA | Mark phase complete, validate transition, advance task chain |\n| `bkit_pdca_plan` | Template | Generate plan document template with level-specific sections and task chain |\n| `bkit_pdca_design` | Template | Generate design template (Starter/Dynamic/Enterprise variants) |\n| `bkit_pdca_analyze` | Template | Generate gap analysis template |\n| `bkit_pdca_next` | PDCA | Recommend next PDCA action based on current state |\n| `bkit_classify_task` | Utility | Classify task size (quick_fix / minor_change / feature / major_feature) |\n| `bkit_detect_level` | Utility | Detect project level from directory structure |\n| `bkit_select_template` | Utility | Select template by phase and level |\n| `bkit_check_deliverables` | Utility | Verify phase deliverables exist |\n| `bkit_memory_read` | Memory | Read session memory |\n| `bkit_memory_write` | Memory | Write session memory |\n\n---\n\n## Skills (27)\n\n| Skill | Category | Trigger Examples |\n|-------|----------|-----------------|\n| **pdca** | Core | `$pdca plan`, `$pdca design`, `$pdca analyze` |\n| **plan-plus** | Core | \"brainstorming plan\", \"deep planning\", \"intent discovery\" |\n| **bkit-rules** | Core | Core rules (auto-applied via AGENTS.md) |\n| **bkit-templates** | Core | \"plan template\", \"design template\" |\n| **starter** | Level | \"static site\", \"portfolio\", \"beginner\" |\n| **dynamic** | Level | \"login\", \"fullstack\", \"authentication\" |\n| **enterprise** | Level | \"microservices\", \"k8s\", \"terraform\" |\n| **development-pipeline** | Pipeline | \"where to start\", \"development order\" |\n| **phase-1-schema** | Pipeline | \"schema\", \"data model\", \"terminology\" |\n| **phase-2-convention** | Pipeline | \"coding rules\", \"conventions\" |\n| **phase-3-mockup** | Pipeline | \"mockup\", \"wireframe\", \"prototype\" |\n| **phase-4-api** | Pipeline | \"API design\", \"REST endpoints\" |\n| **phase-5-design-system** | Pipeline | \"design system\", \"component library\" |\n| **phase-6-ui-integration** | Pipeline | \"frontend integration\", \"API client\" |\n| **phase-7-seo-security** | Pipeline | \"SEO\", \"security hardening\" |\n| **phase-8-review** | Pipeline | \"architecture review\", \"gap analysis\" |\n| **phase-9-deployment** | Pipeline | \"CI/CD\", \"production deployment\" |\n| **code-review** | Quality | \"review code\", \"check quality\" |\n| **zero-script-qa** | Quality | \"test logs\", \"QA without scripts\" |\n| **mobile-app** | Platform | \"React Native\", \"Flutter\", \"iOS app\" |\n| **desktop-app** | Platform | \"Electron\", \"Tauri\", \"desktop app\" |\n| **codex-learning** | Learning | \"learn Codex\", \"Codex CLI setup\" |\n| **bkend-quickstart** | bkend.ai | \"bkend setup\", \"first project\" |\n| **bkend-data** | bkend.ai | \"table\", \"CRUD\", \"schema\", \"filter\" |\n| **bkend-auth** | bkend.ai | \"signup\", \"login\", \"JWT\", \"session\" |\n| **bkend-storage** | bkend.ai | \"file upload\", \"presigned URL\", \"CDN\" |\n| **bkend-cookbook** | bkend.ai | \"tutorial\", \"example project\", \"todo app\" |\n\nSkills use progressive disclosure -- only metadata is loaded initially, with full instructions injected when activated.\n\n---\n\n## Language Support\n\nbkit automatically detects your language from trigger keywords in 8 languages:\n\n| Language | Trigger Keywords |\n|----------|-----------------|\n| English | static website, beginner, API design, verify, analyze |\n| Korean | 정적 웹, 초보자, API 설계, 검증, 분석 |\n| Japanese | 静的サイト, 初心者, API設計, 確認, 分析 |\n| Chinese | 静态网站, 初学者, API设计, 验证, 分析 |\n| Spanish | sitio web estatico, principiante, verificar |\n| French | site web statique, debutant, verifier |\n| German | statische Website, Anfanger, prufen |\n| Italian | sito web statico, principiante, verificare |\n\n---\n\n## Configuration\n\n### MCP Server (`.codex/config.toml`)\n\n```toml\n[mcp_servers.bkit]\ncommand = \"node\"\nargs = [\"./.bkit-codex/packages/mcp-server/index.js\"]\nstartup_timeout_sec = 10\ntool_timeout_sec = 60\nrequired = true\n```\n\n### bkit Config (`bkit.config.json`)\n\n```json\n{\n  \"pdca\": {\n    \"matchRateThreshold\": 90,\n    \"maxIterations\": 5,\n    \"statusFile\": \"docs/.pdca-status.json\",\n    \"memoryFile\": \"docs/.bkit-memory.json\"\n  },\n  \"taskClassification\": {\n    \"thresholds\": { \"quickFix\": 10, \"minorChange\": 50, \"feature\": 200 }\n  },\n  \"conventions\": {\n    \"naming\": {\n      \"components\": \"PascalCase\",\n      \"functions\": \"camelCase\",\n      \"constants\": \"UPPER_SNAKE_CASE\",\n      \"files\": \"kebab-case\"\n    }\n  },\n  \"supportedLanguages\": [\"en\", \"ko\", \"ja\", \"zh\", \"es\", \"fr\", \"de\", \"it\"]\n}\n```\n\n---\n\n## Relationship to bkit-claude-code\n\nbkit-codex is a port of [bkit-claude-code](https://github.com/popup-studio-ai/bkit-claude-code), adapted for OpenAI Codex CLI compatibility. Key differences:\n\n| Aspect | bkit-claude-code | bkit-codex |\n|--------|-----------------|------------|\n| Platform | Claude Code | OpenAI Codex CLI |\n| Automation | Hook-Driven (100% auto) | Instruction-Driven (~70% auto) |\n| Context file | CLAUDE.md | AGENTS.md |\n| Skills format | YAML frontmatter | SKILL.md + openai.yaml |\n| Agent system | 16 dedicated agents | SKILL.md descriptions |\n| State management | Hook scripts | MCP tools |\n| Hooks | 10 hook events (45 scripts) | N/A (rules in AGENTS.md) |\n| Team orchestration | CTO-Led Agent Teams | N/A |\n| Functions | 241 (5 lib modules) | 80 (4 lib modules) |\n| Commands | Slash commands (`/pdca`) | Skill invocation (`$pdca`) |\n| Dependencies | Node.js modules | Zero external dependencies |\n\n### What's Preserved\n\n- 3 Core Principles (Automation First, No Guessing, Docs = Code)\n- PDCA workflow with phase transition validation\n- 3-Level system (Starter / Dynamic / Enterprise)\n- 9-Phase development pipeline\n- 8-language trigger support\n- PDCA document templates\n- State file format (`docs/.pdca-status.json`)\n\n### What's Different\n\n- No hook system -- enforcement relies on AGENTS.md rules + MCP tool guidance\n- No team orchestration -- Codex CLI doesn't support agent spawning\n- Reduced function set (80 of 241) -- excluded hook I/O, team management, Claude Code-specific modules\n- Skills use Codex-native `SKILL.md` format with `openai.yaml` manifest\n- Context recovery mode for state reconstruction after compaction\n- Task chain with automatic phase-to-phase advancement\n\n---\n\n## Documentation\n\n- [Installation Guide](docs/installation.md) -- Detailed setup instructions\n- [Architecture](docs/architecture.md) -- System design and context strategy\n- [Migration Guide](docs/migration-guide.md) -- Migrating from bkit-claude-code\n- [Skills API](docs/api/skills-api.md) -- All 26 skills reference\n- [MCP API](docs/api/mcp-api.md) -- All 16 MCP tools reference\n- [Changelog](CHANGELOG.md) -- Version history\n\n---\n\n## Contributing\n\nWe welcome contributions! Please see [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.\n\n### Quick Contribution Guide\n\n1. Fork the repository\n2. Create a feature branch from `main`\n3. Make changes following [code style guidelines](CONTRIBUTING.md#code-style)\n4. Run tests: `node --test packages/mcp-server/tests/`\n5. Submit a pull request\n\n### Key Rules\n\n- **Zero dependencies**: MCP server must use pure Node.js only\n- **Consistent patterns**: All tools follow `{ handler, definition }` pattern\n- **8-language triggers**: Include triggers in all 8 supported languages\n- **JSDoc comments**: Required for all public functions in `lib/`\n\n### Branch Protection\n\n- Only `admin` team members can merge to `main`\n- All changes require pull request review\n- Version releases are managed through Git tags\n\n---\n\n## License\n\nCopyright 2024-2026 POPUP STUDIO PTE. LTD.\n\nLicensed under the Apache License, Version 2.0. See [LICENSE](LICENSE) for details.\n\n---\n\n## Support\n\n- **Issues**: [GitHub Issues](https://github.com/popup-studio-ai/bkit-codex/issues)\n- **Email**: contact@popupstudio.ai\n\n---\n\nMade with AI by [POPUP STUDIO](https://popupstudio.ai)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpopup-studio-ai%2Fbkit-codex","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpopup-studio-ai%2Fbkit-codex","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpopup-studio-ai%2Fbkit-codex/lists"}