mini-swe-agent
What it is
A minimal Python library for building autonomous software engineering agents that uses only bash commands to interact with codebases. It features a linear message history and supports execution in local or sandboxed environments like Docker.
The problem it solves
Existing AI coding agents are often over-engineered, require complex configurations, and rely on stateful shell sessions that are difficult to sandbox and scale.
What you could use it for
- Automating the resolution of GitHub issues
- Performing software engineering tasks in sandboxed environments
- Serving as a lightweight baseline for evaluating LLM coding capabilities
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.