Introduction
One challenge in open source is not the lack of opportunities, it’s deciding which opportunity to work on next.
Many repositories have hundreds of open issues. Some are beginner-friendly, some require deep project knowledge, and others may no longer be relevant. As contributors, we often spend a significant amount of time reviewing issues before making a contribution.
To explore how automation could help, I built an AI-powered contribution assistant using n8n.
The goal was to automatically collect issues, evaluate them against predefined criteria, prioritize them, and generate a daily contribution digest.
The Problem
When reviewing open source repositories, contributors often face questions such as:
- Which issue matches my current skill set?
- Which issue can be worked on within the time I have available?
- Which issue provides the best learning opportunity?
- Which issue has a high probability of resulting in a merged contribution?
- Which issue should I start with today?
Answering these questions manually across multiple repositories can become time-consuming.
The Solution
The workflow automates issue discovery and prioritization.
It:
- Retrieves open issues from GitHub repositories
- Filters out pull requests
- Collects issue metadata
- Uses an LLM to analyze and rank opportunities
- Generates a daily digest
- Stores recommendations in Notion for future tracking
The objective is not to replace contributor judgment, but to reduce the effort required to identify promising contribution opportunities.
Workflow Overview
The workflow consists of four major stages:
1. Issue Collection
Issues are fetched from GitHub repositories.
Relevant information is extracted, including:
- Labels
- Comment count
- Last update time
- Repository source
2. Issue Evaluation
An LLM evaluates issues against predefined criteria such as:
- Technology relevance
- Documentation relevance
- Learning value
- Expected effort
- Probability of successful contribution
3. Recommendation Generation
The workflow generates:
- Top contribution opportunities
- Quick wins
- Learning-focused opportunities
- A recommended issue to start with
4. Knowledge Management
Recommendations are stored in Notion and delivered via email.
This creates a searchable backlog instead of requiring contributors to revisit repositories repeatedly.
Benefits
Reduced Decision Fatigue
Contributors spend less time searching and more time contributing.
Consistent Engagement
A daily recommendation can encourage regular participation in open source projects.
Personalized Prioritization
Different contributors value different things. The ranking criteria can be adjusted based on individual goals.
Better Visibility
Storing recommendations in Notion makes it easier to track contribution opportunities over time.
Challenges Encountered
- Building the workflow surfaced several practical challenges:
- Prompt design significantly impacted recommendation quality.
- Structured JSON outputs were necessary for reliable automation.
- Notion schema mismatches caused workflow failures.
- LLM-generated outputs required validation before downstream processing.
- Integration testing took considerably longer than expected.
The workflow itself was relatively straightforward. Making the components work reliably together required the most effort.
Possible Improvements
The current workflow works well as a personal assistant, but there are several areas for improvement.
Historical Tracking
Track previously recommended issues and avoid recommending the same issue repeatedly.
Contribution Feedback Loop
Record completed contributions and use that information to improve future recommendations.
Skill-Based Profiles
Allow contributors to define profiles such as:
- Documentation Contributor
- Automation Engineer
- Backend Developer
- Frontend Developer
and generate recommendations accordingly.
Multi-Repository Support
Expand beyond a single ecosystem and evaluate opportunities across multiple projects.
**Effort Estimation
**Use issue history and metadata to estimate actual effort more accurately.
Duplicate Detection
Avoid recommending issues that are very similar or represent the same underlying work.
Weekly and Monthly Reports
Generate summaries that show:
- Issues reviewed
- Contributions completed
- Skills explored
- Areas of growth
Key Takeaway
The most interesting lesson from this experiment was that the bottleneck was not finding open source work, it was selecting the right work.
Automating issue discovery and prioritization helped transform a large list of open issues into a smaller set of actionable recommendations.
While the workflow is still evolving, it demonstrates how automation and AI can be used to reduce overhead and help contributors spend more time learning, building, and contributing.
Top comments (0)