contribs

A visualization of my GitHub contributions across the attentive-mobile repositories, deployed as a static site on GitHub Pages.

Live: https://ixtli.github.io/contribs/ (active once Pages is enabled; see below)

See CLAUDE.md for the development guide (architecture, workflows, and design constraints) when resuming work.

The dashboard is a single-page app that loads SQLite in the browser via WASM (sql.js) and queries the committed dataset directly — no build step, no server, no API token at view time. The data is collected by the contrib-dataset tool.

Current scope: pull requests authored by ixtli in attentive-mobile repos, created this year (since 2026-01-01) — see tools/contrib-dataset/repos.txt for the repo list. The dataset is metadata only — repo, PR number, state, timestamps, line/comment/review counts and decisions. No PR titles, comment/review bodies, or file paths are stored, so the dashboard is safe to serve publicly.

Layout

Path	What
public/index.html	The dashboard SPA. Loads sql.js + Chart.js from a CDN, fetch()es contrib.db, runs the aggregate SQL client-side, and renders KPIs, six charts, and four tables.
tools/contrib-dataset/contrib.db	The committed SQLite dataset the SPA queries. Source of truth.
tools/contrib-dataset/	The data tool: contrib.py syncs PRs / comments / reviews from the GitHub API into SQLite; analyze.py is an offline CLI report (and can emit a self-contained HTML dashboard).
tools/contrib-dataset/repos.txt	The attentive-mobile repo list the sync runs over.
.github/workflows/pages.yml	Assembles the SPA + contrib.db and deploys to GitHub Pages.

Deploy on GitHub Pages

1. In the repo: Settings → Pages → Build and deployment → Source: GitHub Actions.
2. The pages.yml workflow runs on every push to main (and can be run manually via Actions → Deploy dashboard to GitHub Pages → Run workflow). It copies public/index.html and tools/contrib-dataset/contrib.db into the published site and deploys — no Python, no token, no network at deploy time.
3. The dashboard is served at the Pages URL.

Pages on a private repo requires a paid GitHub plan, and the published site is public. That's fine here — the dataset contains no titles/bodies/paths, only counts and trends.

Preview locally

The SPA needs contrib.db served next to it over HTTP (a file:// open won't fetch the DB):

cp tools/contrib-dataset/contrib.db public/
python3 -m http.server -d public 8000   # then open http://localhost:8000

Refresh the data (manual)

Re-sync with a token, then commit the updated DB — the push redeploys Pages:

cd tools/contrib-dataset
export GH_TOKEN=<a GitHub token with repo read scope>
./contrib.py sync --repos @repos.txt --since 2026-01-01   # incremental; safe to re-run
git commit -am "Refresh contribution data" && git push

To see the numbers in the terminal (no browser): ./analyze.py --db contrib.db.

See tools/contrib-dataset/README.md for full tool docs and tools/contrib-dataset/FINDINGS.md for notes from the original (access-limited) first pass.