alibaba/arthas
Arthas is an open-source Java diagnostic tool by Alibaba that enables real-time production troubleshooting and performance monitoring without code changes or JVM restarts.
Awesome AI Agents › Code Automation and Testing
modal-labs/devlooper
Devlooper is an innovative program synthesis agent designed to autonomously generate and iteratively improve code by running tests and fixing errors until all tests pass. It extends the capabilities of the smol developer project by integrating with Modal's sandbox environment, which allows the agent to execute code in isolated containers and incrementally build the environment by installing necessary packages. This approach enables the agent to diagnose and fix issues in the code through a debug loop that involves running tests, analyzing error outputs, and generating a debug plan to modify files, install packages, or run commands to resolve problems. The project supports multiple environment templates for different programming languages and frameworks, including React with Jest, Python, and Rust, with the flexibility to add more templates for other languages. This makes devlooper adaptable to a wide range of development scenarios. The sandbox environment ensures that code execution and testing are done safely and reproducibly, leveraging containerization techniques similar to Docker. Devlooper's debug loop is a key feature where the agent uses outputs from failed tests to inform the next steps in code correction. It separates the diagnosis phase from the action phase, which improves the accuracy of the language model's debugging capabilities. The debug plan can include inspecting and fixing files, installing packages, or running commands within the container. Users can set up devlooper by creating accounts on Modal and OpenAI, installing necessary dependencies, and running the agent with prompts describing the desired software project and selecting an environment template. The output is saved locally once the code passes all tests. The project is a proof of concept with plans for enhancements such as user feedback integration, improved debugging prompts, documentation fetching, loop prevention, and support for more language models. Overall, devlooper represents a significant step towards autonomous software development by combining program synthesis, automated testing, and iterative debugging in a containerized environment.
https://github.com/modal-labs/devlooper
Also in Code Automation and Testing
SWE-agent/SWE-agent
SWE-agent is an autonomous software engineering tool that uses large language models to automatically fix GitHub issues, perform web tasks, and solve cybersecurity challenges, achieving state-of-the-art results in automated software repair and offensive cybersecurity.
qodo-ai/qodo-cover
Qodo-Cover is an AI-powered tool that automates the generation of unit tests to enhance code coverage, supporting integration with GitHub CI workflows and local CLI usage.
judge0/judge0
Judge0 is a robust, scalable, and open-source online code execution system that supports over 60 programming languages and is designed for easy integration into various applications.
BuilderIO/micro-agent
BuilderIO/micro-agent is an AI-powered tool that writes and iteratively refines code by generating tests and ensuring all test cases pass, focusing on reliable test-driven development.
mattzcarey/shippie
Shippie is an AI-powered extensible code review agent that integrates into CI/CD pipelines to automatically detect code issues and improve software delivery speed.
OpenAutoCoder/Agentless
Agentless is an open-source project that automatically solves software development problems through a three-phase process of fault localization, repair, and patch validation, achieving competitive results on SWE-bench benchmarks.
the-open-engine/zeroshot
Command-line orchestrator that drives coding agent CLIs through an executor-verifier loop, so the agent that wrote a change is never the one that declares it correct.