Zero2Repo: Can Coding Agents Build Repositories from Scratch?
cs.SE, cs.AI
Submitted: 2026-09-29
Updated: 2026-10-01
Code: https://github.com/OpenEdgeHQ/Zero2Repo
Project page: https://zero2repo.ai
Sources
- Evaluating Large Language Models Trained on Code
- Program Synthesis with Large Language Models
- NL2Repo-Bench: Towards Long-Horizon Repository Generation Evaluation of Coding Agents
- Prompting Large Language Models to Tackle the Full Software Development Lifecycle: A Case Study
- RepoGenesis: Benchmarking End-to-End Microservice Generation from Readme to Repository
- RepoZero: Can LLMs Generate a Code Repository from Scratch?
- Multi-SWE-bench: A Multilingual Benchmark for Issue Resolving
- Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces
- SWE-Together: Evaluating Coding Agents in Interactive User Sessions
- Automated Benchmark Generation for Repository-Level Coding Tasks
- TerminalWorld: Benchmarking Agents on Real-World Terminal Tasks
Related papers
- Falsification-Based Verification of LLM-Generated Optimization Models: Sound Test Batteries and Their Detection Limits
- GitSkills: A Dataset of Agent Skills on GitHub
- SABER: Benchmarking Operational Safety of LLM Coding Agents in Stateful Project Workspaces
- PackMonitor: Enabling Zero Package Hallucinations Through Decoding-Time Monitoring
- IntentCoding: Amplifying User Intent in Code Generation
- Incentives and Outcomes in Bug Bounties