This content originally appeared on DEV Community and was authored by Emiliano Antonini
"Hello and thank you. I'd kindly ask you to test my latest project: a tool that teaches you while you write code. It would be great if you could leave your thoughts in the comments."
Code Analysis System
Overview
This is a comprehensive code analysis system built with React, TypeScript, and Express, offering multi-agent capabilities for examining code across various dimensions: security vulnerabilities, performance bottlenecks, syntax issues, and structural concerns. It provides a modern web interface with real-time analysis results and detailed reporting. The business vision is to create a learning-focused code analysis tool that evolves by learning from effective fixes, automatically applying them in the future to create increasingly shorter and more stable patterns. The project's ambition is to be an open-source, community-benefiting tool, not a commercial product.
User Preferences
Preferred communication style: Simple, everyday language. App sharing: Use direct demo link (https://code-master-ai-raptus2012.replit.app/demo) for sharing instead of root URL to showcase features immediately. Tool Vision: Learning-focused code analysis tool that creates increasingly shorter and more stable patterns from saved fix combinations. The goal is to learn from effective fixes and apply them automatically in the future. Project Goals: Open source release with optional donations - built for community benefit, not commercial sale. Layout preference: 50%/50% horizontal split with code fixed on left, results scrolling on right. UI Enhancement: Completed Windows-like movable AI chat with drag, resize, minimize/maximize, and full interaction capabilities (completed January 6, 2025). Chat window features transparent resize handles, smooth drag functionality, and proper minimize behavior.
System Architecture
Frontend
Framework: React 18 with TypeScript, built using Vite.
Styling: Tailwind CSS with shadcn/ui and Radix UI primitives.
State Management: TanStack Query for server state.
Routing: Wouter for client-side routing.
UI Components: CodeInput (editor with syntax highlighting), AnalysisResults (comprehensive display), AgentResults (individual agent reports), ComparativeAnalysis (prioritized recommendations).
Design: Modern card-based layout, consistent typography, dark/light mode support, gradient backgrounds, and professional visual polish.
Backend
Runtime: Node.js with Express.js.
Language: TypeScript with ES modules.
Database: PostgreSQL with Drizzle ORM (Neon serverless).
Session Storage: PostgreSQL-based.
Development: Hot reload with Vite middleware.
Code Analysis Engine
Multi-Agent System: Seven specialized agents: Security, Performance, Syntax, Structural, Dependency, Refactoring, and Code Quality.
Language Support: Java, JavaScript, Python, C++, C#, PHP, Go, Rust, Pascal, Kotlin, Swift, Ruby, Scala, Orial (custom Italian language), Oracle SQL, Oracle PL/SQL.
Analysis Storage: Results stored in PostgreSQL with JSON fields.
Pattern Learning: Automatic pattern registration upon analysis, collective evolution model with simulation for success rates, and a universal learning database across all supported languages, including Orial, Oracle SQL, and Oracle PL/SQL.
Dynamic Features: Automatic language detection, dynamic agent display based on user settings with auto-save, and "Go to line" navigation for results.
Fix Strategy: Focuses on suggestion-only mode; automatic code modification is disabled to prevent structural issues, with future auto-fixes relying on learned patterns.
AI Chat: Full multilingua support with interactive chat assistant that can teach, explain, and analyze code in all 16+ supported languages including C++, Java, Python, Orial, Oracle SQL, Oracle PL/SQL, etc.
External Dependencies
Database: u/neondatabase/serverless, drizzle-orm, drizzle-kit.
Frontend Libraries: u/tanstack/react-query, u/radix-ui/*, tailwindcss, lucide-react, class-variance-authority.
Backend Libraries: express, zod.
Development Tools: vite, typescript, tsx.
API Integrations: OpenAI API (for AI chat and pattern learning).
This content originally appeared on DEV Community and was authored by Emiliano Antonini
Emiliano Antonini | Sciencx (2025-08-09T20:34:57+00:00) CodeCraft. Retrieved from https://www.scien.cx/2025/08/09/codecraft/
Please log in to upload a file.
There are no updates yet.
Click the Upload button above to add an update.