onirudda-islam / stats

Auto-generated language statistics surface for @onirudda-islam's GitHub profile.

All source repositories are private. This repo exists as a public read surface — a weekly GitHub Action scans all private repos, aggregates language byte counts, and commits the results here.

Files

File	Description
output/languages.json	Full language breakdown — bytes + percentages
output/summary.json	Top 10 languages + repo count metadata
output/languages.svg	Stacked bar chart SVG for embedding in README

Embedding in your profile README

![Language Stats](https://raw.githubusercontent.com/onirudda-islam/stats/main/output/languages.svg)

How it works

GitHub Action (weekly cron)
  └─ Python script (scripts/aggregate_languages.py)
       └─ GitHub API /user/repos  →  all repos (public + private)
       └─ GitHub API /repos/:owner/:repo/languages  →  per-repo bytes
       └─ Aggregate + sort + generate SVG + JSON
       └─ git commit → this repo (main branch)

Setup

1. Create a fine-grained PAT with:
   • repo scope (to read private repo languages)
   • contents: write on THIS repo (to push the output)
2. Add it as a secret named STATS_GH_TOKEN in this repo's settings
3. Run the workflow manually once to verify, then let the weekly schedule take over

---

Last generated: see output/summary.json