All 229 problem statements from Smart India Hackathon 2026 in one searchable place - clean markdown, JSON and CSV, plus a fast bilingual web app to browse, filter and shortlist them.
| Path | What it is |
|---|---|
ps_2026/ |
229 markdown files (SIH26001.md ... SIH26229.md) + index |
data/sih2026_ps.json |
Full structured export (all fields incl. descriptions) |
data/sih2026_ps.csv |
Same data, spreadsheet-friendly |
data/changelog/ |
Field-level diffs of every daily data update |
CHANGELOG.md |
Rolling record of what changed in the dataset |
scripts/scrape_sih.py |
Reproducible scraper + validator |
web/ |
Next.js + shadcn/ui web app (search, filters, shortlist) |
A fast, fully static, SEO-optimized browsing experience in English and Hindi:
- Fuzzy search across title, description, organization, theme and PS number
(⌘K / Ctrl+K command palette,
/focuses search) - Filters by theme, category (Software/Hardware), organization, dataset availability
- Detail pages with deadline countdowns, similar statements, share on WhatsApp, copy as Markdown, open-in-chat and private notes
- Shortlist (stored in your browser) with CSV/Markdown export and WhatsApp sharing
- Branded OG cards for every statement when shared on social media
- Dark mode, keyboard navigation, mobile-first, fully accessible
cd web
npm install
npm run dev # http://localhost:3000
npm run build # fully static, 461 prerendered pages
npm run lint| Metric | Value |
|---|---|
| Total problem statements | 229 |
| Software | 175 |
| Hardware | 54 |
| Themes | 17 |
| Organizations | 30 |
| Source | sih.gov.in/sih2026PS |
A GitHub Actions workflow re-scrapes sih.gov.in every 24 hours (04:00 IST) and automatically:
- Updates the markdown, JSON and CSV exports with the latest deadlines, submitted-idea counts, dataset links and any new/changed statements
- Writes a field-level diff of what changed to
data/changelog/<date>.mdand prepends it to CHANGELOG.md (added/removed/updated statements with old -> new values per field) - Validates the result (record count sanity check + field validation) - if the scrape looks incomplete, nothing is committed
- Commits and pushes only when the data actually changed, so there are no noisy commits
Every PR and push is also checked by the CI workflow (dataset validation + lint + typecheck + build).
python3 scripts/scrape_sih.py # fetch live + regenerate everything
python3 scripts/scrape_sih.py --validate # validate existing artifacts
python3 scripts/scrape_sih.py --cache FILE.html # parse a cached copy offlineRequires beautifulsoup4 + lxml.
- Dataset (problem statement content): CC BY 4.0 - content is published by Smart India Hackathon / Innovation Cell, Government of India. Attribution: "Source: Smart India Hackathon, sih.gov.in".
- Code (scraper, web app): MIT.
Note: Deadlines and submitted-idea counts are point-in-time snapshots and may change on sih.gov.in. Always verify on the official portal before submitting.
Found incorrect data or have an idea? See CONTRIBUTING.md and the issue templates.
This is an unofficial, community-maintained archive. It is not affiliated with, endorsed by, or sponsored by Smart India Hackathon or the Government of India.