mini-swe-agent · SWE-agent team (Princeton/Stanford)

confidence: highactive5,636

A deliberately radical ~100-line Python agent with bash as its only tool (no tool-calling API, no config sprawl) that still scores >74% on SWE-bench Verified, used by Meta, NVIDIA, and IBM as a baseline/benchmarking harness and proof that scaffolding can be minimal.

Homepagehttps://mini-swe-agent.com
Repositoryhttps://github.com/SWE-agent/mini-swe-agent
Licenseopen-source
Domaincoding
Form factorcli
Model-agnostic✓ yes
Native models
Local models✓ yes
Pricingbyok
MCP client○ no
Agent Skills support○ no
Plugins / hooks○ no
Sandboxingoptional
Autonomy spectrumsupervised → autonomous-local
Subagents○ no
Parallel sessions✓ yes
Context managementnone-observed
Headless / API✓ yes
GitHub integration✓ yes
SWE-bench Verified74%
Sources

✓ profile last verified 2026-07-08 · benchmark figures are vendor-reported unless a source states otherwise