OSS Opportunity Radar

Find open-source projects with strong demand but weak commercialization — the gaps where a commercial product could thrive.

OSS Opportunity Radar is an automated tool that mines GitHub to surface open-source opportunities: projects where there is clearly unmet demand (discovered from real issue/discussion signals) but little or no organized commercial offering yet. Every opportunity ships with a data-backed insight — how many users are asking for what, and whether anyone is already monetizing it.

Live demo: https://wushidiguo.github.io/opportunity-radar/

---

Motivation

Most "open-source opportunity" lists rank projects by stars. This is misleading: high popularity usually means the project is already commercialized — e.g. Supabase, ComfyUI, n8n and their hosts all have product sites, cloud offerings and pricing pages. Ranking by fame finds the worst opportunities, not the best.

This project scores on the commercialization gap instead: reward projects with real, verified demand that nobody has nicely packaged for money yet.

What it finds

• Strong demand: measured from open issues, activity and usage signals, not just stars.
• Commercialization gap: a project site, cloud/hosted keywords, or a huge star count push a project down, because those are signs it is already monetized.
• Real insights: the engine reads open issue text and the README to count what users actually ask for — hosted/cloud versions, paying, mobile, API/integration, features — and whether the maintainers are leaving those requests unanswered.

Features

• Web app — a static, zero-build opportunity browser: switch topics, search, sort, filter, save favorites, with an insight panel per card.
• Scoring engine (v2) — popularity + demand + activity + health + commercialization gap.
• Insight engine — mines open issues + README to count demand signals (hosted / paid / mobile / API / feature / self-host), unanswered requests, and detect monetization — then writes a one-line, data-backed reason.
• CLI — table / JSON / CSV output, min-score filter, offline demo mode, list/content filtering.
• CI/CD — tests on every push, a daily data snapshot, and automatic GitHub Pages deploys.

Live demo

The live site is refreshed automatically every day by a scheduled GitHub Actions job:

https://wushidiguo.github.io/opportunity-radar/

How it works

1. Collect — search GitHub for candidate repositories per topic (a larger pool than just the top stars).
2. Score — run the v2 scoring engine. Unlike naive popularity scoring, it penalizes known-commercialized projects (product site, cloud/hosted keywords, huge stars) and rewards real demand plus a commercialization gap.
3. Mine insight — for the top candidates, read their open issues and README to quantify demand signals and detect monetization evidence, producing a concrete "why someone would pay" reason.
4. Publish — rank the opportunities and serve them in the web app (rebuilt daily).

Opportunity score (0-100)

Factor	Max	Description
popularity	20	log10(stars+1) x 5 (sublinear, so big projects do not dominate)
demand	15	open_issues / 150
activity	10	pushed within 180d = 10; within 365d = 4
health	15	not archived + has issues + has description + has wiki
gap	40 (can be negative)	commercialization gap (core)

Gap signals: a non-GitHub product homepage, or cloud/saas/enterprise/managed keywords in the description, mark a project as commercialized and drag it down. No homepage, a DIY/self-hosted topic, or a permissive license push it up. Very large star counts (>120k) are penalized as likely-commoditized. Grades: A >= 75, B >= 60, C >= 45, D < 45.

Getting started

Prerequisites

• Node.js >= 18
• GitHub CLI (gh) authenticated — needed for live data. Use --demo mode to run fully offline.

CLI

# Offline demo (no gh needed)
node bin/opportunity-radar.mjs ai 10 --demo

# Live data (gh must be logged in)
node bin/opportunity-radar.mjs ai 10

# JSON export + min-score filter
node bin/opportunity-radar.mjs database 20 --format json --min-score 70

# CSV export (one column per factor)
node bin/opportunity-radar.mjs self-hosted 20 --format csv --out opportunities.csv

Build the data snapshot

node scripts/build-snapshot.mjs --topics ai,self-hosted,database,devops,llm --pool 120 --top 40

This fetches a larger candidate pool, scores it, mines insights for the top candidates, and writes web/data/opportunities.json for the web app.

Run the web app locally

node scripts/serve.mjs
# open http://localhost:4173

Run the tests

npm test

Project structure

opportunity-radar/
  bin/opportunity-radar.mjs   # CLI entry
  src/
    score.mjs                 # v2 scoring engine (pure, explainable)
    insight.mjs               # insight engine (issue + README mining)
    fetch.mjs                 # GitHub data fetching / fixture fallback
    render.mjs                # table / json / csv rendering
    cli.mjs                   # arg parsing + main flow
  scripts/
    build-snapshot.mjs        # build the web data snapshot
    serve.mjs                 # local static server
  web/                        # static web app (no build step)
  test/                       # unit / integration / render / insight tests
  .github/workflows/          # CI + daily snapshot + Pages deploy

Contributing

Contributions are welcome. Please open an issue to discuss a change before opening a pull request, and make sure npm test passes locally.

License

MIT

Acknowledgments

Inspired by the observation that open-source popularity and commercialization are often inversely related when it comes to opportunity. The demand signals come directly from public GitHub issue data.