SWE-agent/mini-swe-agent

The 100 line AI agent that solves GitHub issues or helps you in your command line. Radically simple, no huge configs, no giant monorepo—but scores >74% on SWE-bench verified!

87
/ 100
Verified

Executes each bash command independently via `subprocess.run` rather than maintaining shell state, enabling straightforward sandboxing and scaling across docker/podman/singularity/bubblewrap. Routes any LLM through litellm/openrouter/portkey for model flexibility, supporting completion and response endpoints with interleaved thinking. Maintains a completely linear message history with no special processors, making it ideal for debugging, fine-tuning, and understanding the exact prompts sent to the model.

3,292 stars and 1,358,879 monthly downloads. Used by 1 other package. Actively maintained with 13 commits in the last 30 days. Available on PyPI.

Maintenance 20 / 25
Adoption 21 / 25
Maturity 24 / 25
Community 22 / 25

How are scores calculated?

Stars

3,292

Forks

455

Language

Python

License

MIT

Last pushed

Mar 12, 2026

Monthly downloads

1,358,879

Commits (30d)

13

Dependencies

14

Reverse dependents

1

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/agents/SWE-agent/mini-swe-agent"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.