Match your code with purpose.
Demo Video: YouTube
GitIsMatch is an intelligent web application that bridges the gap between developers and open-source contributions. By analyzing your GitHub profile using Google's Gemini AI, it creates a unique "coding persona" based on your repositories and bio, then actively searches for "help wanted" issues that align with your specific skills and interests.
- AI-Powered Profile Analysis: Utilizes Google Gemini 2.5 Flash to read your bio, top repositories, languages, and topics to create a comprehensive summary of your expertise.
- Intelligent Issue Discovery: Automatically generates complex, optimized GitHub search queries based on your analysis to find relevant issues.
- Smart Filtering & Refinement:
- Filter results instantly by clicking on AI-identified skill tags.
- Add Custom Tags: Manually add specific technologies you want to work with.
- Deep Search: Select tags and trigger a refined search to query GitHub specifically for those topics.
- High-Volume Results: Fetches up to 100 relevant issues per search to ensure you have plenty of options.
- Privacy Focused: Runs entirely in the browser. Your API token (if provided) is never stored permanently.
- Frontend: React 19, TypeScript, Vite
- Styling: Tailwind CSS
- AI: Google Gemini API (
@google/genai) - Data: GitHub REST API
- Icons: Lucide React
- Node.js installed.
- A Google Gemini API Key.
- (Optional) A GitHub Personal Access Token for higher rate limits.
- Clone the repository.
- Install dependencies:
npm install
- Set up your environment variables (Gemini API Key).
- Start the development server:
npm run dev
- Enter a Profile: Paste a GitHub URL (e.g.,
https://github.com/torvalds) or username. - Analyze: The app fetches public data and sends it to Gemini for analysis.
- Explore: View your "Match Report" containing a summary of your skills and a list of curated issues.
- Refine:
- Click tags to filter the current list.
- Click "+ Add" to input new skills (e.g., "Rust").
- Select multiple tags and click "Search GitHub for Selected Tags" to perform a fresh, targeted search.
This project is licensed under the MIT License.