SWE-agent
swe-agent.com
- Free plan
- Windows
- macOS
- Linux
- Web
- Self-hosted
8.0 Club score Visit swe-agent.com
What the maker publishes
- Coding agent
- Yes source
- Browser setup
- SWE-agent can be run in a browser-based VS Code environment, where users provide language-model API keys in a .env file. source
- Browser use
- The setup guide offers a browser-based VS Code environment for a quick first look and lists Node.js as a requirement for the web-based GUI. source
- Configuration
- The project says SWE-agent is configurable and fully documented, with configuration governed by a single YAML file. source
- Configuration
- The project describes itself as configurable through a YAML file and designed to be simple and hackable for research. source
- Credentials
- Users must provide the relevant model keys and, for private GitHub repositories, GitHub credentials; keys can be set as environment variables, in a .env file, or through a command-line option. source
- Credentials
- Users supply their own language-model keys and need a GitHub key to access private repositories. source
- Current recommendation
- The maintainers recommend mini-swe-agent, which they say matches SWE-agent’s performance while being simpler. source
- Customization
- Users can add custom tools and change the demonstration trajectory supplied to the agent at the start of each run. source
- Images
- SWE-agent supports vision-capable models for processing images from GitHub issues, and the FAQ says images must be under 10 MB. source
- Image support
- The documentation describes multimodal support for processing images from GitHub issues and says images must be under 10 MB. source
- Installation
- The preferred setup is installation from source using pip. source
- Local model limits
- For local models, the documentation says cost tracking cannot be used unless a custom model registry supplies cost information. source
- Models
- The documentation says SWE-agent supports models served through LiteLLM and local models with an OpenAI-compatible API. source
- Model support
- The documentation says SWE-agent supports models available through LiteLLM and local models served through an OpenAI-compatible API. source
- Operating systems
- The FAQ says SWE-agent runs on Windows, macOS, and Linux, with Docker container availability possibly limiting environment choices. source
- Platforms
- The FAQ says SWE-agent runs on Windows, macOS, and Linux, with Docker environment availability as a possible limitation. source
- Project status
- The documentation says SWE-agent has been superseded by mini-swe-agent and is now in maintenance-only mode. source
- Purpose
- SWE-agent enables a language model to use tools autonomously to fix issues in GitHub repositories, find cybersecurity vulnerabilities, or perform custom tasks. source
- Purpose
- SWE-agent uses language models to autonomously work on software engineering tasks, including fixing issues in GitHub repositories. source
- Research design
- Its Agent-Computer Interface provides commands and feedback for models to browse repositories, view and edit files, and run code. source
- Research focus
- The project describes itself as made for research and simple and hackable by design. source
- Run records
- The FAQ says `.traj` output files contain complete records of SWE-agent's thought process and actions. source
- Runtime
- Docker is the default backend; the documentation says cloud code evaluation is also possible without Docker. source
Checked against the maker's pages on 12 Oct 2026.