mini-swe-agent
Minimal open-source software-engineering agent from the SWE-bench/SWE-agent team: about 100 lines of Python, bash as its only tool, a linear history, any model via LiteLLM.
Tags
Description
Each action runs independently with subprocess.run, so swapping in docker exec or another sandbox is trivial. Runs locally or in Docker/Podman, Singularity/Apptainer, bubblewrap and more. The README reports >74% on SWE-bench Verified, and it is the scaffold behind the SWE-bench (bash only) leaderboard. From the team behind SWE-agent and SWE-bench. Installs from PyPI as mini-swe-agent.
Related
- SWE-bench: Benchmark and leaderboards (Verified, Lite, Multimodal, Multilingual) testing whether LLM agents can resolve real-world GitHub issues.
- SWE-smith: Toolkit and datasets for scaling training data for software-engineering agents by synthesizing task instances from GitHub repos (NeurIPS 2025 D&B).
- OpenHands: Open-source, model-agnostic platform for cloud coding agents.
- Aider: Open-source AI pair programming in your terminal.
Guides & comparisons
- Agent benchmarks compared (Comparison): Ten widely cited agent benchmarks (SWE-bench, Terminal-Bench, tau-bench, BFCL, MCPMark, GAIA, BrowseComp, WebArena, Mind2Web and OSWorld) compared on what they measure, how many tasks they have, where the agent runs, how runs are scored, where results are published and how they are licensed.
Sources
- https://github.com/SWE-agent/mini-swe-agent
- https://mini-swe-agent.com/latest/
- https://pypi.org/project/mini-swe-agent/
Machine-readable
- JSON
- Markdown
- Source file on GitHub (edit via pull request)