Super Analyzer is an application that identifies and evaluates common anti-patterns in C++, Java, Python, and Rust code. It uses a multi-agent architecture with three coordinated agent types—Primary, Fixer, and Chat—to deliver accurate analysis and iterative improvements. The Web UI supports natural multi-turn interactions and uses a PostgreSQL database to persist in user accounts and conversation history across sessions. Super Analyzer also provides a REST/API layer and a Python interface, all secured through user/password authentication backed by PostgreSQL. Access is protected with JWTs signed using RSA-256, ensuring a consistent and secure experience across all entry points.

• Using Reasoning models. The framework leverages Nvidia Nemotron 3 Super reasoning model for heavy analysis tasks.
• Actor critic pattern to fix code. Using the reasoning model in an actor critic pattern we achieve coding results similar to those from dedicated coding models.
• Chatbot for Analysis. A chatbot style web UI can be used for analysis and fixing code.
• Multiple Endpoints. The app exposes a Python SDK and a REST API in addition to the web interface.
• Authentication built in. The app uses user / password authentication for all end points. Credentials are managed with RSA 256 + JWT tokens.

To learn about the app architecture, implementation details and example datasets:

• https://huggingface.co/blog/balaatdell/super-analyzer-app-on-dell-enterprise-hub/