CLAPP: The CLASS LLM Agent for Pair Programming

Santiago Casas, Christian Fidler, Boris Bolliet, Francisco Villaescusa-Navarro, Julien Lesgourgues
8/7/2025
astro-ph.IMastro-ph.COcs.AI

No official code found. Best implementation: santiagocasas/clapp (14 stars, MIT, last commit 2026-02-11).

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

View on arXivView PDF

Code Implementations(1)

santiagocasas/clappCommunity72%
MIT

CLAPP: CLASS LLM Agent for Pair Programming

146Shell, PythonMar 25, 20258 months agoMIT
ag2agentic-aiaiclasscosmology+2 more

Similar Papers

What is the Role of Large Language Models in the Evolution of Astronomy Research?

83% match2024

AI Assistants to Enhance and Exploit the PETSc Knowledge Base

83% match2025

Scaling Large Vision-Language Models for Enhanced Multimodal Comprehension In Biomedical Image Analysis

82% match2025

SciSciGPT: Advancing Human-AI Collaboration in the Science of Science

82% match2025

Bridging Literature and the Universe Via A Multi-Agent Large Language Model System

82% match1 repo52025

Designing an Evaluation Framework for Large Language Models in Astronomy Research

82% match1 repo22024

Cite this paper

@article{casas2025clapp,
  title  = {CLAPP: The CLASS LLM Agent for Pair Programming},
  author = {Santiago Casas and Christian Fidler and Boris Bolliet and Francisco Villaescusa-Navarro and Julien Lesgourgues},
  year   = {2025},
  eprint = {2508.05728},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2508.05728v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2508.05728.svg)](https://codeofpaper.com/paper/2508.05728)

Discussion