Imagen de Portada

RepoLaunch: Automating Build&Test Pipeline of Code Repositories on ANY Language and ANY Platform

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Li, Kenan, Li, Rongzhi, Zhang, Linghao, Jin, Qirui, Zhu, Liao, Huang, Xiaosong, Zhang, Geng, Zhang, Yikai, He, Shilin, Xie, Chengxing, Zhang, Xin, Jin, Zijian, Li, Bowen, Zhang, Chaoyun, Kang, Yu, Huang, Yufan, Nallipogu, Elsie, Rajmohan, Saravan, Lin, Qingwei, Zhang, Dongmei
Formato:	Preprint
Publicado:	2026
Materias:	Software Engineering Machine Learning Multiagent Systems I.2.5; I.2.7
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

_version_	1866917316016472064
author	Li, Kenan Li, Rongzhi Zhang, Linghao Jin, Qirui Zhu, Liao Huang, Xiaosong Zhang, Geng Zhang, Yikai He, Shilin Xie, Chengxing Zhang, Xin Jin, Zijian Li, Bowen Zhang, Chaoyun Kang, Yu Huang, Yufan Nallipogu, Elsie Rajmohan, Saravan Lin, Qingwei Zhang, Dongmei
author_facet	Li, Kenan Li, Rongzhi Zhang, Linghao Jin, Qirui Zhu, Liao Huang, Xiaosong Zhang, Geng Zhang, Yikai He, Shilin Xie, Chengxing Zhang, Xin Jin, Zijian Li, Bowen Zhang, Chaoyun Kang, Yu Huang, Yufan Nallipogu, Elsie Rajmohan, Saravan Lin, Qingwei Zhang, Dongmei
contents	Building software repositories typically requires significant manual effort. Recent advances in large language model (LLM) agents have accelerated automation in software engineering (SWE). We introduce RepoLaunch, the first agent capable of automatically resolving dependencies, compiling source code, and extracting test results for repositories across arbitrary programming languages and operating systems. To demonstrate its utility, we further propose a fully automated pipeline for SWE dataset creation, where task design is the only human intervention. RepoLaunch automates the remaining steps, enabling scalable benchmarking and training of coding agents and LLMs. Notably, several works on agentic benchmarking and training have recently adopted RepoLaunch for automated task generation.
format	Preprint
id	arxiv_https___arxiv_org_abs_2603_05026
institution	arXiv
publishDate	2026
record_format	arxiv
spellingShingle	RepoLaunch: Automating Build&Test Pipeline of Code Repositories on ANY Language and ANY Platform Li, Kenan Li, Rongzhi Zhang, Linghao Jin, Qirui Zhu, Liao Huang, Xiaosong Zhang, Geng Zhang, Yikai He, Shilin Xie, Chengxing Zhang, Xin Jin, Zijian Li, Bowen Zhang, Chaoyun Kang, Yu Huang, Yufan Nallipogu, Elsie Rajmohan, Saravan Lin, Qingwei Zhang, Dongmei Software Engineering Machine Learning Multiagent Systems I.2.5; I.2.7 Building software repositories typically requires significant manual effort. Recent advances in large language model (LLM) agents have accelerated automation in software engineering (SWE). We introduce RepoLaunch, the first agent capable of automatically resolving dependencies, compiling source code, and extracting test results for repositories across arbitrary programming languages and operating systems. To demonstrate its utility, we further propose a fully automated pipeline for SWE dataset creation, where task design is the only human intervention. RepoLaunch automates the remaining steps, enabling scalable benchmarking and training of coding agents and LLMs. Notably, several works on agentic benchmarking and training have recently adopted RepoLaunch for automated task generation.
title	RepoLaunch: Automating Build&Test Pipeline of Code Repositories on ANY Language and ANY Platform
topic	Software Engineering Machine Learning Multiagent Systems I.2.5; I.2.7
url	https://arxiv.org/abs/2603.05026