{
  "markdown": "# AI Development Standards\n\n[![CI Status](https://img.shields.io/badge/CI-passing-brightgreen)](https://github.com/daffy0208/ai-dev-standards/actions)\n[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)\n[![TypeScript](https://img.shields.io/badge/TypeScript-5.0+-blue)](https://www.typescriptlang.org/)\n[![Node](https://img.shields.io/badge/Node-20+-green)](https://nodejs.org/)\n\n**Version 3.1.0** | Last Updated: 2025-11-24\n\n> A comprehensive framework of specialized AI skills, MCP servers, and development tools for AI-assisted software development. Features automated validation, agent evaluation, and quality assurance.\n\n## 📊 Current Resources\n\n**Total: 199 Resources**\n\n| Category         | Count | Description                                       |\n| ---------------- | ----- | ------------------------------------------------- |\n| **Skills**       | 64    | Specialized AI methodologies and workflows        |\n| **MCPs**         | 51    | Model Context Protocol servers (executable tools) |\n| **Tools**        | 4     | Core utility scripts                              |\n| **Components**   | 75    | Reusable UI and system components                 |\n| **Integrations** | 5     | Third-party service connectors                    |\n\n**MCP Coverage:** 79.7% (51 MCPs supporting 64 Skills)\n\n---\n\n## ✨ What's New in 3.1.0\n\n### 🎯 Agent Evaluation System (Phase 5.12)\n\nImplement Eval-Driven Development (EDD) for continuous agent quality assurance:\n\n- Automated testing against golden datasets\n- Multiple grading strategies (exact, regex, LLM-based)\n- Performance metrics and regression tracking\n- 100% test pass rate in validation suite\n\n```bash\n# Run agent evaluations\nnode scripts/run-agent-evals.js --dataset tests/fixtures/golden-dataset-example.json --mock\n```\n\n### ⚡ Two-Tier Validation System\n\n- **Quick Validation** (10-30s): Registry consistency, documentation\n- **Full Validation** (2-5min): Includes linting, type checking, tests, agent evals\n\n```bash\nnpm run validate:quick  # Fast feedback\nnpm run validate:full   # Comprehensive checks\n```\n\n### 📚 Enhanced Documentation\n\n- `.claude/CLAUDE.md` - Complete Claude Code configuration guide\n- `FINAL-RESOURCE-COUNTS.md` - Resource tracking and metrics\n- `docs/VALIDATION-SYSTEM.md` - Validation methodology\n\n---\n\n## 🚀 Quick Start\n\n> 📖 **New to this repository?** Check out our [Installation Guide](INSTALL.md) and [Quick Start Guide](QUICK-START-GUIDE.md) for step-by-step instructions.\n\n### For New Projects\n\n```bash\n# Clone the repository\ngit clone https://github.com/daffy0208/ai-dev-standards.git\ncd ai-dev-standards\n\n# Install dependencies\nnpm install\n\n# Run validation to ensure everything works\nnpm run validate\n```\n\n### For Existing Projects\n\n```bash\n# Clone as a reference\ngit clone https://github.com/daffy0208/ai-dev-standards.git ~/ai-dev-standards\n\n# Reference skills and patterns in your .cursorrules or .claude/claude.md\n# See docs/EXISTING-PROJECTS.md for integration guide\n```\n\n### Using with Claude Code\n\n1. Open your project in Claude Code\n2. Reference this repository in your project instructions:\n\n   ```markdown\n   You have access to ai-dev-standards at ~/ai-dev-standards\n\n   When needed, reference skills from skills/ and patterns from standards/\n   Use the skill-registry.json to find relevant skills for tasks\n   ```\n\n3. Claude will automatically discover and use appropriate skills\n\n---\n\n## 📖 What This Repository Does\n\nThink of this as **a shared knowledge base** between you and Claude:\n\n### 🎓 64 Specialized Skills\n\nMethodologies Claude follows automatically:\n\n- **Product**: mvp-builder, product-strategist, go-to-market-planner\n- **AI/ML**: rag-implementer, multi-agent-architect, knowledge-graph-builder\n- **Development**: frontend-builder, api-designer, backend-architect\n- **Infrastructure**: deployment-advisor, security-engineer, performance-optimizer\n- **Design**: ux-designer, visual-designer, design-system-architect\n- **Quality**: testing-strategist, quality-auditor, agent-evaluator\n\n### 🔧 51 MCP Servers\n\nExecutable tools that extend Claude's capabilities:\n\n- **Search**: semantic-search-mcp, dark-matter-analyzer-mcp\n- **Quality**: code-quality-scanner-mcp, security-scanner-mcp, test-runner-mcp\n- **AI/Data**: vector-database-mcp, embedding-generator-mcp, knowledge-base-mcp\n- **Design**: figma-sync-mcp, design-token-manager-mcp, theme-builder-mcp\n- **DevOps**: deployment-orchestrator-mcp, database-migration-mcp\n\n### 📐 Architecture Patterns\n\nProven approaches for complex systems:\n\n- RAG architectures (Naive, Advanced, Modular)\n- Multi-agent coordination patterns\n- Event-driven systems\n- Real-time data pipelines\n- Authentication patterns\n\n### 🛡️ Quality Assurance\n\n- Automated validation system (2-tier)\n- Agent evaluation framework (EDD)\n- Security best practices\n- Performance standards\n- Accessibility guidelines\n\n---\n\n## 💡 Key Features\n\n### ⚡ Automated Validation\n\n```bash\n# Quick validation (10-30 seconds)\nnpm run validate:quick\n\n# Full validation (2-5 minutes)\nnpm run validate:full\n\n# Agent evaluation only\nnode scripts/run-agent-evals.js --dataset tests/fixtures/golden-dataset-example.json --mock\n```\n\n**Validates:**\n\n- ✅ Registry consistency\n- ✅ Documentation accuracy\n- ✅ Code quality (ESLint)\n- ✅ Type safety (TypeScript)\n- ✅ Test coverage\n- ✅ **Agent performance (NEW)**\n\n### 🤖 Agent Evaluation System\n\nTest AI agents against golden datasets to ensure consistent, high-quality outputs:\n\n```javascript\n{\n  \"tests\": [\n    {\n      \"id\": \"T001\",\n      \"input\": \"Create a React button component with TypeScript\",\n      \"expected\": \"import React from 'react';\",\n      \"grading\": { \"type\": \"contains\", \"threshold\": 0.8 }\n    }\n  ]\n}\n```\n\n**Features:**\n\n- Multiple grading types (exact match, contains, regex, LLM-graded)\n- Performance metrics (latency, success rate, score)\n- Historical tracking and regression detection\n- Custom dataset support\n\n### 📊 Comprehensive Documentation\n\n- **For Developers**: `docs/GETTING-STARTED.md`, `docs/QUICK-START.md`\n- **For AI**: `meta/PROJECT-CONTEXT.md`, `meta/HOW-TO-USE.md`\n- **Configuration**: `.claude/CLAUDE.md`, `FINAL-RESOURCE-COUNTS.md`\n- **Validation**: `docs/VALIDATION-SYSTEM.md`\n\n### 🎯 Smart Resource Discovery\n\n```bash\n# Find skills for a task\ngrep -r \"mvp\" meta/skill-registry.json\n\n# Search all resources\ngrep -r \"authentication\" meta/\n\n# View resource counts\ncat FINAL-RESOURCE-COUNTS.md\n```\n\n---\n\n## 🗂️ Repository Structure\n\n```\nai-dev-standards/\n├── skills/                    # 64 specialized methodologies\n│   ├── mvp-builder/          # MVP development & prioritization\n│   ├── rag-implementer/      # RAG system implementation\n│   ├── api-designer/         # API design patterns\n│   └── [61 more...]\n│\n├── mcp-servers/              # 51 executable tools\n│   ├── semantic-search-mcp/  # Semantic code search\n│   ├── vector-database-mcp/  # Vector DB integration\n│   ├── code-quality-scanner-mcp/\n│   └── [48 more...]\n│\n├── standards/                # Architecture & best practices\n│   ├── architecture-patterns/\n│   ├── best-practices/\n│   ├── coding-conventions/\n│   └── project-structure/\n│\n├── meta/                     # Resource registry & context\n│   ├── registry.json         # Master resource registry\n│   ├── skill-registry.json   # Skill catalog\n│   ├── mcp-registry.json     # MCP catalog\n│   └── PROJECT-CONTEXT.md    # For AI assistants\n│\n├── docs/                     # Comprehensive documentation\n│   ├── GETTING-STARTED.md\n│   ├── VALIDATION-SYSTEM.md\n│   ├── AGENT-VALIDATION.md   # NEW!\n│   └── [40+ more guides...]\n│\n├── scripts/                  # Automation & validation\n│   ├── run-agent-evals.js    # NEW! Agent evaluation\n│   ├── validate-full.sh      # Full validation suite\n│   └── [20+ more scripts...]\n│\n└── tests/                    # Test suites & fixtures\n    ├── fixtures/\n    │   └── golden-dataset-example.json  # NEW!\n    └── [150+ test files...]\n```\n\n---\n\n## 🎯 Usage Examples\n\n### Example 1: Starting a New Project\n\n```\nUser: \"I want to build a SaaS product for invoice management\"\n\nClaude uses:\n1. product-strategist → Validate problem-solution fit\n2. mvp-builder → Identify P0 features (invoicing, payment tracking)\n3. frontend-builder → React/Next.js structure\n4. api-designer → REST API design\n5. deployment-advisor → Vercel + Railway recommendation\n6. security-engineer → Auth, data encryption, PCI compliance\n```\n\n### Example 2: Implementing AI Search\n\n```\nUser: \"Add AI-powered search to our documentation\"\n\nClaude uses:\n1. rag-implementer → RAG methodology\n2. rag-pattern.md → Advanced RAG architecture\n3. vector-database-mcp → Pinecone integration\n4. embedding-generator-mcp → OpenAI embeddings\n5. semantic-search-mcp → Search implementation\n```\n\n### Example 3: Code Quality Audit\n\n```\nUser: \"Audit our codebase for quality issues\"\n\nClaude uses:\n1. quality-auditor → Comprehensive audit methodology\n2. code-quality-scanner-mcp → Static analysis\n3. security-scanner-mcp → Vulnerability detection\n4. performance-profiler-mcp → Performance bottlenecks\n5. test-runner-mcp → Test coverage analysis\n6. agent-evaluator → AI agent quality checks (NEW!)\n```\n\n---\n\n## 🔍 Finding Skills\n\n### By Task\n\n```bash\n# Search skills by keyword\ngrep -i \"authentication\" meta/skill-registry.json\ngrep -i \"database\" meta/skill-registry.json\ngrep -i \"testing\" meta/skill-registry.json\n```\n\n### By Category\n\nView `meta/skill-registry.json` for complete categorization:\n\n- **Product & Business** (8 skills)\n- **AI & Machine Learning** (10 skills)\n- **Frontend Development** (6 skills)\n- **Backend Development** (8 skills)\n- **Infrastructure & DevOps** (8 skills)\n- **Design & UX** (12 skills)\n- **Quality & Testing** (12 skills)\n\n### Auto-Discovery\n\nSkills activate automatically based on your conversation with Claude. Just describe what you want to build!\n\n---\n\n## ⚙️ Validation System\n\n### Two-Tier Approach\n\n#### Tier 1: Quick Validation (10-30 seconds)\n\n```bash\nnpm run validate:quick\n```\n\n**Checks:**\n\n- Registry consistency\n- Documentation accuracy\n- Configuration files\n- Basic CLI functionality\n\n**Use when:** Before commits, during rapid development\n\n#### Tier 2: Full Validation (2-5 minutes)\n\n```bash\nnpm run validate:full\n```\n\n**Checks:**\n\n- Everything in Tier 1 +\n- ESLint code quality\n- TypeScript type checking\n- Unit & integration tests\n- **Agent Evaluation (Phase 5.12)** ✨ NEW\n- Build verification\n\n**Use when:** Before pushing, in CI/CD, before releases\n\n### Agent Evaluation (Phase 5.12)\n\nTest AI agents against golden datasets:\n\n```bash\n# Run with mock agent (for testing)\nnode scripts/run-agent-evals.js --dataset tests/fixtures/golden-dataset-example.json --mock\n\n# Run with real agent (production)\nnode scripts/run-agent-evals.js --dataset tests/fixtures/golden-dataset-example.json\n\n# Verbose output\nnode scripts/run-agent-evals.js --dataset tests/fixtures/golden-dataset-example.json --mock --verbose\n```\n\n**Output:**\n\n```\n📊 Summary\n----------------------------------------\nTotal Tests:    10\nPassed:         10\nFailed:         0\nPass Rate:      100.0%\nAvg Score:      0.96\nAvg Latency:    47ms\n----------------------------------------\n\n✅ Agent Evaluations PASSED\n```\n\nSee `docs/VALIDATION-SYSTEM.md` for complete methodology.\n\n---\n\n## 📚 Documentation\n\n### Getting Started\n\n- `docs/QUICK-START.md` - 5-minute quick start\n- `docs/GETTING-STARTED.md` - Comprehensive setup guide\n- `docs/EXISTING-PROJECTS.md` - Integration for existing projects\n\n### Validation & Quality\n\n- `docs/VALIDATION-SYSTEM.md` - Validation methodology\n- `docs/AGENT-VALIDATION.md` - Agent evaluation guide (NEW!)\n- `.claude/commands/validate.md` - Validation command reference\n\n### Configuration\n\n- `.claude/CLAUDE.md` - Claude Code configuration (NEW!)\n- `FINAL-RESOURCE-COUNTS.md` - Resource metrics (NEW!)\n- `meta/PROJECT-CONTEXT.md` - For AI assistants\n- `meta/HOW-TO-USE.md` - Navigation guide\n\n### Development\n\n- `CONTRIBUTING.md` - Contribution guidelines\n- `docs/MCP-DEVELOPMENT-ROADMAP.md` - MCP development guide\n- `docs/TROUBLESHOOTING.md` - Common issues\n\n---\n\n## 🛠️ Development\n\n### Running Tests\n\n```bash\n# Run all tests\nnpm test\n\n# Run specific test suites\nnpm run test:unit          # Unit tests only\nnpm run test:registry      # Registry validation\nnpm run test:cli           # CLI tests\n\n# Run agent evaluations\nnpm run test:agent-eval    # Agent evaluation suite\n```\n\n### Validation Commands\n\n```bash\n# Linting\nnpm run lint               # Check code quality\nnpm run lint:fix           # Auto-fix issues\n\n# Type Checking\nnpm run typecheck          # TypeScript validation\n\n# Formatting\nnpm run format             # Format code with Prettier\nnpm run format:check       # Check formatting\n\n# Registry\nnpm run validate:registries  # Validate resource registries\nnpm run generate:registries  # Regenerate registries\n```\n\n### Creating Custom Datasets\n\nCreate your own agent evaluation datasets:\n\n```json\n{\n  \"version\": \"1.0.0\",\n  \"description\": \"Your custom test dataset\",\n  \"tests\": [\n    {\n      \"id\": \"T001\",\n      \"category\": \"code-generation\",\n      \"description\": \"Test description\",\n      \"input\": \"Your test prompt\",\n      \"expected\": \"Expected output or pattern\",\n      \"grading\": {\n        \"type\": \"contains\", // or \"exact\", \"regex\", \"llm-graded\"\n        \"threshold\": 0.8\n      },\n      \"tags\": [\"category\", \"feature\"]\n    }\n  ]\n}\n```\n\n---\n\n## 📊 Quality Metrics\n\n### Resource Coverage\n\n- **Skills**: 64 specialized methodologies\n- **MCPs**: 51 executable tools\n- **MCP Coverage**: 79.7% (51 MCPs / 64 Skills)\n- **Documentation**: 100% of skills documented\n\n### Validation Status\n\n- ✅ **Registry Validation**: Passing\n- ✅ **Type Checking**: Passing\n- ✅ **Linting**: Passing (790 warnings, 0 errors)\n- ✅ **Agent Evaluation**: Passing (100% success rate)\n- ✅ **Test Coverage**: 78%\n\n### Performance\n\n- **Agent Evaluation**: 47ms avg latency\n- **Quick Validation**: 10-30 seconds\n- **Full Validation**: 2-5 minutes\n\n---\n\n## 🗺️ Roadmap\n\n### ✅ Completed\n\n- **v3.1.0** (2025-11-24): Agent Evaluation System\n  - Phase 5.12 implementation\n  - Golden dataset support\n  - Multiple grading strategies\n  - Performance metrics\n\n- **v3.0.3** (2025-11-14): Validation System\n  - Two-tier validation\n  - Registry automation\n  - Documentation consolidation\n\n- **v2.1.0** (2025-10-29): Orchestration\n  - Claude Code integration\n  - Registry validation\n  - 100% resource discovery\n\n### 🔜 Planned\n\n- **v3.2.0**: Enhanced Agent Evaluation\n  - Real agent integration\n  - Advanced LLM grading\n  - Regression tracking dashboard\n\n- **v3.3.0**: MCP Expansion\n  - Additional development MCPs\n  - Better skill-MCP coverage\n  - Performance improvements\n\n- **v4.0.0**: Ecosystem Integration\n  - GitHub Actions workflows\n  - VSCode extension\n  - Web dashboard\n\n---\n\n## 🤝 Contributing\n\nWe welcome contributions! See `CONTRIBUTING.md` for guidelines.\n\n### Ways to Contribute\n\n1. **Add Skills**: Create new specialized methodologies\n2. **Add MCPs**: Build executable tools\n3. **Improve Documentation**: Clarify guides and examples\n4. **Report Issues**: Help us find and fix bugs\n5. **Create Datasets**: Expand agent evaluation coverage\n\n### Development Setup\n\n```bash\n# Clone the repository\ngit clone https://github.com/daffy0208/ai-dev-standards.git\ncd ai-dev-standards\n\n# Install dependencies\nnpm install\n\n# Run validation\nnpm run validate:quick\n\n# Make changes and test\nnpm test\n\n# Submit PR\n```\n\n---\n\n## 📝 License\n\nMIT License - see [LICENSE](LICENSE) for details\n\n---\n\n## 🙏 Acknowledgments\n\nThis repository synthesizes best practices from:\n\n- Claude Code official patterns\n- Production software development\n- AI-assisted development research\n- Community feedback and contributions\n\n**Maintained by:** [@daffy0208](https://github.com/daffy0208)\n\n---\n\n## 📞 Support\n\n- **Documentation**: `docs/` directory\n- **Issues**: [GitHub Issues](https://github.com/daffy0208/ai-dev-standards/issues)\n- **Discussions**: [GitHub Discussions](https://github.com/daffy0208/ai-dev-standards/discussions)\n\n---\n\n## 🔗 Quick Links\n\n### For Developers\n\n- [Quick Start](docs/QUICK-START.md)\n- [Getting Started](docs/GETTING-STARTED.md)\n- [Validation System](docs/VALIDATION-SYSTEM.md)\n- [Troubleshooting](docs/TROUBLESHOOTING.md)\n\n### For AI Assistants\n\n- [Project Context](meta/PROJECT-CONTEXT.md)\n- [How to Use](meta/HOW-TO-USE.md)\n- [Skill Registry](meta/skill-registry.json)\n- [MCP Registry](meta/mcp-registry.json)\n\n### Configuration\n\n- [Claude Code Config](.claude/CLAUDE.md)\n- [Resource Counts](FINAL-RESOURCE-COUNTS.md)\n- [Validation Command](.claude/commands/validate.md)\n\n---\n\n**Built for excellence in AI-assisted development** 🚀\n",
  "bytes": 16569,
  "sha": "5095d5e8e2a9714bb8449171a0bb8b781638b86916eacb3248efeb01f1b70837",
  "repo_slug": "daffy0208/ai-dev-standards",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/skl_daffy0208_ai_dev_standards_animation_des_10c91ebf/readme"
}