CLAPP: The CLASS LLM Agent for Pair Programming

TL;DR

CLAPP is an AI-powered interactive assistant that enhances research with the CLASS cosmology solver by providing conversational coding support, debugging, and visualization through LLMs and domain-specific retrieval.

Contribution

This paper presents CLAPP, a novel multi-agent LLM system integrating semantic search and live execution to assist researchers in computational cosmology tasks.

Findings

Improves coding efficiency for CLASS users

Reduces entry barrier for non-expert scientists

Enhances human-AI collaboration in cosmology research

Abstract

We introduce CLAPP (CLASS LLM Agent for Pair Programming), an interactive AI assistant designed to support researchers working with the Einstein-Boltzmann solver CLASS. CLAPP leverages large language models (LLMs) and domain-specific retrieval to provide conversational coding support for CLASS-answering questions, generating code, debugging errors, and producing plots. Its architecture combines multi-agent LLM orchestration, semantic search across CLASS documentation, and a live Python execution environment. Deployed as a user-friendly web application, CLAPP lowers the entry barrier for scientists unfamiliar with AI tools and enables more productive human-AI collaboration in computational and numerical cosmology. The app is available at https://classclapp.streamlit.app