JobScout Lite

Copyright © 2026 Noumaan Nabi. All rights reserved.

Free-tier job intelligence dashboard. See PROGRESS.md for build status and updates.md for the full changelog — this repo is being built step by step.

Auth note: email confirmation is off

This project turns off Supabase's "Confirm email" setting (Authentication → Providers → Email → Confirm email → off). Signing up logs you in immediately — no confirmation email. This sidesteps Supabase's free-tier cap of 2 auth emails/hour on the built-in mailer, which isn't worth fighting for a low-stakes personal scraper tool. If you ever want it back (e.g. before sharing this with other people), flip that toggle back on — the signup code already handles both cases gracefully, it just won't show a "check your email" screen since it doesn't need to.

Password reset still uses email (Supabase requires it to prove you own the account) — but that's rare enough in normal use that the 2/hour cap essentially never bites there.

Setup

.env.local is already filled in with your Supabase project — nothing to configure. Just:

npm install
npm run dev

Open http://localhost:3000, sign up / log in (see Step 2 notes below for the email-confirmation flow), and you'll land on the dashboard.

The scraper (Step 5)

Located in scraper/. Sources: RemoteOK (public JSON API), We Work Remotely (public RSS feeds), Greenhouse + Lever (public per-company job board APIs — you list which companies to track), and Wellfound (best-effort HTML — see the caveat in scraper/sources/wellfound.py, it may return 0 results since Wellfound needs JS rendering).

Run it once, manually

cd scraper
pip install -r requirements.txt --break-system-packages
python main.py

This creates scraper/jobs.db (SQLite — the 4-day job cache), writes scraper/logs/scraper.log, and overwrites cache/jobs.json (which /api/jobs/cache serves to the frontend) with whatever it found.

Track specific companies on Greenhouse / Lever

Edit scraper/config.py — GREENHOUSE_BOARD_TOKENS and LEVER_COMPANY_SLUGS. Find a company's token/slug from their public board URL: boards.greenhouse.io/<token> or jobs.lever.co/<slug>.

Schedule it (once you have the Oracle Cloud VPS — later step)

0 */6 * * * /home/ubuntu/jobscout/venv/bin/python /home/ubuntu/jobscout/scraper/main.py

API routes (Step 3)

All under /api, all require a logged-in session except the jobs cache:

• GET /api/jobs/cache — public, returns cache/jobs.json (currently seeded placeholder data; the scraper will overwrite this file in Step 5). Cached 10 minutes.
• GET/POST /api/applications, PATCH/DELETE /api/applications/[id]
• GET/POST /api/saved-jobs, PATCH/DELETE /api/saved-jobs/[id]
• GET/POST /api/blocklist, DELETE /api/blocklist/[id]

All of these are thin wrappers around Supabase — RLS does the actual per-user data isolation; the routes just return clean 401/400/404s.

First-time Supabase setup (already done if you followed Step 2)

1. Free project at supabase.com.
2. Run supabase/schema.sql in the SQL Editor once.
3. .env.local needs NEXT_PUBLIC_SUPABASE_URL and NEXT_PUBLIC_SUPABASE_ANON_KEY from Project Settings → API.

Full deployment instructions (Vercel + Oracle Cloud free VPS + cron) will be added once the scraper and remaining pages exist.