Connect agents to live web environments evaluation.

Agentless🐱: an agentless approach to automatically solve software development problems

Convert Chinese Traditional EPUB book to Simplified

🚀CodiumAI PR-Agent: An AI-Powered 🤖 Tool for Automated Pull Request Analysis, Feedback, Suggestions and More! 💻🔍

Artifact of "Searching for Multi-Fault Programs in Defects4J", SSBSE 2021

We Have No Moat, And Neither Does Devin

CrossCodeEval: A Diverse and Multilingual Benchmark for Cross-File Code Completion (NeurIPS 2023)

DeepSeek Coder: Let the Code Write Itself

Grep source code and see useful code context about matching lines

Harness used to benchmark aider against SWE Bench benchmarks

aider is AI pair programming in your terminal

Collections of research, benchmarks and tools towards more robust and reliable language models for code; LM4Code; LM4SE; reliable LLM; LLM4Code

Replication package of a paper "Large Language Models are Few-shot Testers: Exploring LLM-based General Bug Reproduction"

RepairLLaMA: Efficient Representations and Fine-Tuned Adapters for Program Repair

CodiumAI Cover-Agent: An AI-Powered Tool for Automated Test Generation and Code Coverage Enhancement! 💻🤖🧪🐞

RepairLLaMA tool for the APR competition

LDB: A Large Language Model Debugger via Verifying Runtime Execution Step by Step

VulTrigger is a tool to for identifying vulnerability-triggering statements across functions and investigating the effectiveness of function-level vulnerability detectors in detecting inter-procedu…

Agent fixing SWE bench issues

Artifact for "A Learning-Based Approach to Static Program Slicing": Paper accepted at OOPSLA'24

The source code of project "LLift" (Enhancing static analysis with LLM)

Official Repo for ICML 2024 paper "Executable Code Actions Elicit Better LLM Agents" by Xingyao Wang, Yangyi Chen, Lifan Yuan, Yizhe Zhang, Yunzhu Li, Hao Peng, Heng Ji.