Repo0 framework enables AI to build entire software projects from requirements

Researchers have introduced Repo0, a novel framework designed for zero-to-all code generation, enabling AI agents to construct entire software projects from natural-language requirements. This system maintains an explicit architectural state using a Dual-Directed-Acyclic-Graph (Dual-DAG) to manage component boundaries and guide test-driven development. Evaluations using GPT-5 mini and DeepSeek V3.2 on real-world repositories demonstrated Repo0's superior performance in functionality coverage and pass rates compared to existing baselines. AI

IMPACT This framework could significantly advance the capabilities of AI agents in complex software development tasks.

RANK_REASON This is a research paper detailing a new framework for code generation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX Code Finder for Papers
• CORE Recommender
• DagsHub
• DeepSeek V3.2
• Dual-DAG
• Dual-Directed-Acyclic-Graph
• Gotit.pub
• GPT-5 mini
• Hugging Face
• Influence Flower
• Repo0
• RepoCraft
• ScienceCast

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →