TFM aInventory (2026 Edition)
A unified, offline-first Inventory Management System built as a Progressive Web App (PWA). Features include AI-powered label extraction (OCR), local barcode/QR scanning, and multi-user authentication with LDAP support.
🚀 Quick Start (Production)
For production environments, Docker is the recommended deployment method:
git clone <repository-url> tfm-inventory
cd tfm-inventory
# [D-08] Configuration - Copy examples to actual config files
for f in config/*.yaml.example; do cp "$f" "${f%.example}"; done
# Edit config/secrets.yaml with your JWT_SECRET_KEY and AI keys
nano config/secrets.yaml
# Deploy using the new Python deployment script
python3 scripts/deploy.py production
- Frontend: http://localhost:3000 (or https://localhost:8919 via proxy)
- Backend API: http://localhost:8000/docs
For detailed configuration reference, see config/README.md. For detailed deployment instructions (Docker vs Standalone), see DEPLOYMENT.md.
🏗 Technical Overview
- Backend: FastAPI (Python 3.12+)
- Frontend: Next.js 15+ (React PWA) with responsive Tailwind CSS
- Database: SQLite (SQLAlchemy) with Dexie.js (IndexedDB) for client-side sync.
- AI Engine: Google Gemini (Primary) & Anthropic Claude (Fallback).
- Configuration: [D-07] Consolidated YAML-based config in
config/directory.
For more details on system logic, see PROJECT_ARCHITECTURE.md.
🔐 Security & Constraints
AI agents and developers MUST strictly follow the rules defined in AI_RULES.md.
- No Uppercase UI: All labels/headers must be normal case.
- No Bold Fonts: Text hierarchy is achieved through size and color.
- Offline-First: All features must function without an active network connection.
📦 Production Distribution
To generate a clean production package:
python3 scripts/save_version.py --patch (or --minor/--major)
Last Updated: 2026-04-25
Version: 1.14.25 (Phase 7.1)