SlothContributor: a local agent that does my issue hunting

Description and use

SlothContributor is a local CLI agent that watches chosen repositories, automatically analyzes new issues with a local Gemma 2 model, and delivers a structured "Contributor's Hunch" to save time on research and focus on coding.

Problem

Contributing to open source is hampered by the time spent refreshing GitHub, filtering labels, reading long issue threads, and guessing which files caused a bug, often leading to fatigue before writing a fix.

Outcome

The agent reduced manual research time by eliminating 30–45 minutes of context‑switching, enabling contributors to focus on coding. It also helped the author decide quickly whether an issue was solvable, leading to three merged PRs in the target repository.

Workflow

• 1. Start the watcher with python main.py --watch. 2. The agent polls each repo every 15 minutes. 3. On detecting a new issue, it reads the issue text. 4. Gemma suggests likely file paths. 5. The Python backend verifies paths via the GitHub API. 6. If a file exists, Gemma analyzes the real code and issue together. 7. The agent outputs a "Contributor's Hunch" with core problem, target files, and a suggested fix. 8. The user reviews the hunch and proceeds to implement the fix.

Evidence

I want to contribute to open source, but I hate the hustle. I used to spend 30 minutes refreshing GitHub, filtering labels, reading long issue threads and guessing which files caused a bug.

Source

SlothContributor automates the entire research phase. The second a new issue drops, it doesn't just notify you. It reads the issue, fetches the relevant source code from the repository, and hands you a structured "Contributor's Hunch" on exactly where to look and how to fix it.

Source

The core is a Python CLI that talks to Ollama, running Google's open-weight Gemma 2 (2B) locally.

Source

I added MongoDB Atlas for persistent memory (no duplicate analysis, resumable sessions), Plyer for desktop notifications and Rich for the terminal UI.

Source

Now the hunch gives me a summary of the issue, where to look and a suggested fix up front. I can tell quickly whether an issue is one I understand and can actually solve, and only then ask to be assigned, instead of leaving a vague comment and figuring it out afterwards.

Source