zhangutah/LLM-reasoning-agents: ASE artifact
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
This repository provides a framework for evaluating and deploying Large Language Model (LLM) agents in automated fuzzing tasks. The system generates fuzz harnesses for target functions using LLM-based reasoning, compiles them in a Docker-based OSS-Fuzz environment, and iteratively fixes errors until successful execution. The framework supports C/C++ and Java projects and includes: Automated harness generation using LLM agents Iterative compilation and error fixing Integration with OSS-Fuzz workflows Evaluation pipeline for coverage and fuzzing performance Code is structured into modular components, including agent pipelines, code retrieval tools, and fuzzing utilities.
Fields of Study & Subjects
Software Testing and Debugging Techniques
Software Engineering Research
Software System Performance and Reliability
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 30 March 2026
- Language
- EN