Skip to main content

Analyze Your GitHub Repository

This guide shows you how to build a powerful analytics dashboard for your own GitHub repository using Rill. In just a few commands, you'll have a fully interactive dashboard to discover hot zones in your codebase, track contributor activity, analyze code churn, and measure development velocity.

See it live

Explore the live demo → to see interactive dashboards analyzing real repositories (DuckDB, Rill). This is exactly what you'll create for your own repository.

Step 1: Clone the Project​

Clone from GitHub​

# Clone the GitHub Analytics project
git clone https://github.com/rilldata/rill-examples.git
cd rill-examples/rill-github-analytics

Install Dependencies​

The project includes Python scripts for downloading GitHub data and generating Rill project files. The project uses Poetry for dependency management (installation guide).

# Using Poetry (recommended)
poetry install

# Or using pip with venv
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install pandas pydriller

Step 2: Scrape Git History and Save to Object Storage​

Prerequisites:

The project includes a download_commits.py script that clones your target repository, extracts commit metadata and file changes, then uploads the data as parquet files to your GCS bucket.

# Download and upload to GCS
python download_commits.py owner/repo --gcs --bucket gs://your-bucket/github-analytics

# Or limit to recent commits for faster testing
python download_commits.py owner/repo --gcs --bucket gs://your-bucket/github-analytics --limit 1000

Note: Files will be saved to gs://your-bucket/github-analytics/owner/repo/ to keep data organized by repository.

Private repositories

For private repos, use a fine-grained access token:

  1. Create a fine-grained personal access token with read-only access to the repository
  2. Store it as an environment variable: export GITHUB_TOKEN=your_token_here
  3. Git will automatically use it when cloning (works in local dev and CI)

Note: For large repositories with 10,000+ commits, the download may take 10-30 minutes. Use --limit to test with a smaller dataset first.

Step 3: Generate Rill Project Files​

The project includes a generate_project.py script that will generate:

  • Source definitions pointing to your GCS bucket
  • Data transformation models
  • Metrics definitions
  • An explore dashboard

Run the script with your repository and bucket:

# Generate Rill files configured for your GCS bucket
python generate_project.py owner/repo --gcs --bucket gs://your-bucket/github-analytics

# Examples:
python generate_project.py duckdb/duckdb --gcs --bucket gs://your-bucket/github-analytics
python generate_project.py your-org/your-repo --gcs --bucket gs://your-bucket/github-analytics

Note: Rill supports both Google Cloud Storage (GCS) and Amazon S3. The download script currently supports GCS. For S3, you'll need to modify the script.

Just want to explore locally?

Use the --local flag instead: python generate_project.py owner/repo --local

This is great for testing, but you won't be able to deploy to Rill Cloud without migrating to cloud storage later.

Step 4: Deploy to Rill Cloud​

Deploy your dashboard to share with your team:

rill deploy

This creates a live, shareable link where your team can explore the data together. Since your data is in cloud storage, Rill Cloud can access it directly.

Preview locally first

Want to verify everything looks good before deploying? Run rill start to preview the dashboard locally, then deploy when ready.

Next Steps​

Now that you have GitHub Analytics deployed:

  1. Keep data fresh – Schedule the download script to run regularly (cron, GitHub Actions, etc.) and keep your dashboards up to date
  2. Customize metrics – Edit the metrics YAML files to add team-specific calculations
  3. Add alerts – Use Rill's alerting features to monitor key metrics