FastAPI MCP

AI-first Model Context Protocol server for FastAPI backend development with Smart Blueprints.

🤖 For AI Assistants

CRITICAL: Use context loading prompts for new chats:

1. Choose prompt from .ai-docs/@prompts/@prompt-[task-type].md
2. Copy complete prompt and paste in new chat
3. Wait for ✅ confirmation before proceeding

Quick Reference: Always load .ai-docs/@core/@conventions.md FIRST Validation: python scripts/validate_ai_context.py --check-all

🚀 Quick Start

# Windows: setup.bat && start.bat
# Linux/Mac: ./setup.sh && ./start.sh
# Manual: python -m venv .venv && .venv/Scripts/activate && pip install mcp anyio pydantic && cd frontend-mcp && python -m py_server.server

📁 AI-Optimized Structure

FastAPI-MCP/
├── .ai-docs/@core/        # 🤖 AI CONTEXT (load first)
│   ├── @conventions.md    # Master rules
│   ├── @testing-protocol.md # Testing rules
│   └── @project-concept.md # AI philosophy
├── backend-mcp/           # FastAPI MCP server + smart blueprints
│   ├── blueprints/        # Smart blueprints with embedded templates
│   ├── templates/         # Jinja2 templates (if needed)
│   └── main.py           # FastAPI MCP server
├── tests/                 # Testing infrastructure (unit/integration/load/security)
├── scripts/               # Automation (validate_*.py, smart_blueprint_processor.py)
└── docs/                  # Human documentation

🔧 MCP Tools

• generate_from_smart_blueprint - FastAPI routes, services, models
• smart_crud_route - Complete CRUD API endpoints
• smart_auth_route - Authentication endpoints
• smart_service - Business logic services
• smart_model - Pydantic and SQLAlchemy models

📖 Documentation

For AI: .ai-docs/@core/ (optimized context) For Humans: docs/ (comprehensive guides) For Testing: tests/README.md (current state)