μLearn Contributions Data

Data repository for the μLearn contributions leaderboard, deployed at contributions.mulearn.org.

This repo holds muLearn's config, branding, and scraped contribution data. The app engine (ohcnetwork/leaderboard) is pulled at build time — it is not forked. Only this repo, the scraper plugin (gtech-mulearn/mulearn-github-plugin), the Cloudflare Pages project, and the domain belong to muLearn.

What's here

Path	Purpose
config.yaml	Org info, roles, plugin config, branding pointers
theme.css	μLearn brand colors (overrides the engine's CSS variables)
assets/	Logo, favicon, OG image, badge SVGs (see assets/README.md)
.github/workflows/scrape.yml	Cron (every 6h): scrape the org → commit data → upload DB to R2
.github/workflows/deploy.yml	On push: build the static site → deploy to Cloudflare Pages
contributors/	Generated + human-editable contributor profiles
activities/, aggregates/	Generated scrape output (committed by the workflow)

How it runs

1. scrape.yml runs every 6 hours: authenticates as the mulearn-leaderboard-bot GitHub App, scrapes every repo in gtech-mulearn via the plugin, commits the updated data back to main, and uploads the SQLite DB to Cloudflare R2.
2. The data commit triggers deploy.yml, which builds the static site from the committed data and deploys it to Cloudflare Pages.

Required GitHub Actions secrets / variables

Set under Settings → Secrets and variables → Actions:

Secrets: GH_APP_PRIVATE_KEY (PKCS#8), CLOUDFLARE_API_TOKEN, CLOUDFLARE_ACCOUNT_ID Variables: GH_APP_ID, CLOUDFLARE_PAGES_PROJECT_NAME, LEADERBOARD_DATA_DB_URL

Also enable Settings → Actions → General → Workflow permissions → Read and write.

Assign a role

Everyone defaults to contributor. To mark someone core team, add contributors/<github-username>.md:

---
username: someusername
role: core
---

Commit → the next deploy shows the new role. Human-edited profiles always win over scraped data.