LLM-Powered Git Bisect: Automated Broken Commit Detection with Local AI Explanation

View on GitHub

Pipeline Architecture

When regressions appear after dozens of commits, engineers spend most time on mechanical narrowing before they can even start root-cause analysis. This tool automates that first phase and then summarizes the likely failure mechanism.

It combines deterministic git bisect execution with optional local-model analysis so code and diffs do not leave your machine.

What Makes It Useful

It is not just a bisect wrapper. The output includes structured context around the first bad commit, test command behavior, and an explanation pass that can speed up triage handoff in teams.

If Ollama is unavailable, the bisect still runs and returns actionable commit boundaries.

Run the Project

git clone https://github.com/dakshjain-1616/LLM-Powered-Git-Bisect
cd LLM-Powered-Git-Bisect
pip install -e .
ollama pull qwen3:8b

git-bisect-ai run --good HEAD~30 --bad HEAD --test 'pytest tests/'
git-bisect-ai status --repo .
git-bisect-ai install-alias
git bisect-ai run --good HEAD~30 --bad HEAD --test 'pytest tests/'

The primary command is git-bisect-ai run, and the optional alias flow allows git bisect-ai ... for native Git ergonomics.

Architecture Walkthrough

The llm powered git bisect repository is organized around a clear pipeline, so you can trace the full flow from input handling to final output without guesswork. This makes onboarding easier for new contributors and helps teams debug faster when behavior changes after updates.

Practical Use Cases

If you are evaluating llm powered git bisect for production, start with a small real-world dataset, run the included commands end to end, and compare output quality, latency, and operational complexity. This gives a practical signal that is stronger than a toy demo.

Implementation Notes

The project is useful as both a standalone tool and a reference implementation. You can copy patterns from this codebase into your own stack, especially around evaluation discipline, reproducibility, and operator visibility.

Try NEO in Your IDE

Install the NEO extension to bring AI-powered development directly into your workflow: