SWE-agent
SWE-agent enables language models to autonomously use tools to fix issues in real GitHub repositories, find cybersecurity vulnerabilities, or perform custom tasks. It is an open-source project developed by researchers from Princeton University and Stanford University. Although superseded by mini-swe-agent, it remains in maintenance mode and offers a powerful platform for automated software engineering.
- Automated bug fixing
- Code repository analysis
- Software engineering research
Similar Systems
Systems with similar features, category, or use cases
CAMEL-AI
DeveloperCAMEL-AI is an open-source community dedicated to finding the scaling laws of agents for data generation, world simulation, and task automation. It is the world's first multi-agent system, designed to be data-driven, stateful, and agent-friendly. It enables the creation of collaborative AI agents for complex tasks.
OpenHands
DeveloperOpenHands is an open platform for cloud coding agents, designed to automate real engineering work securely and transparently. It is open-source, model-agnostic, and enterprise-ready, allowing for scaling from one to thousands of agents. The platform reduces engineering toil, speeds up code reviews, expands test coverage, and automates documentation and release notes.
CrewAI
DeveloperStructured framework for agent teams with role specialization. Clear object structure and good memory concept.
Pydantic AI
DeveloperPython-native framework with minimal abstraction. Feels like regular Python development.
We use cookies and similar technologies to analyze the use of our website and improve your experience. Privacy Policy