NetaWorth

Public ledger of wealth declared by India’s elected representatives — sitting MLAs, Lok Sabha MPs, Rajya Sabha MPs, winners, and candidates from reviewed election sources.

Data comes from election affidavits (ECI / ADR / MyNeta). Figures are self-declared, not independently audited market wealth.

Live: netaworth.vercel.app

Prerequisites

• Node.js 22.x
• npm (use npm ci for reproducible installs)

Quick start

npm ci
npm run dev

The app does not require a database for the public UI.

Architecture

Client-rendered Next.js App Router app. Live data is versioned JSON under public/data/:

File	Contents
adr-sitting-mlas-2025.json	Nationwide sitting-MLA snapshot
lok-sabha-sitting-mps.json	Lok Sabha MP snapshot (latest GE winners)
rajya-sabha-sitting-mps.json	Rajya Sabha MP snapshot (ADR March 2026 PDF)
adr-recontest-history.json	Affidavit-to-affidavit comparisons
adr-winner-archive.json	Assembly constituency winners
lok-sabha-winner-archive.json	Lok Sabha winners (2004–2024)
candidates/index.json (+ shards)	Candidate archive

data/election-manifest.json is the reviewed source-coverage input for the candidate, winner, and recontest generators (135 included MyNeta election folders). Counts describe reviewed/imported coverage, not an exhaustive ECI census.

The browser reads these files directly. Changes to db/, drizzle/, or app/api/ do not affect the public UI unless it is migrated to those endpoints.

Optional D1

Cloudflare D1 and Drizzle are an unfinished path. Schema, migrations, importer, and read APIs exist for development, but there is no production database, binding, or UI integration. Do not treat D1 routes as production-ready.

Verification

npm ci
npm run verify

Runs ESLint, a production build, and dataset/UI smoke tests. Separately:

npm run lint
npm run typecheck
npm run build
node --test tests/rendered-html.test.mjs

CI uses Node 22 on pushes to main and on pull requests.

Deployment

Target	URL	Command
Vercel (primary)	https://netaworth.vercel.app	npm run build:vercel
GitHub Pages	https://ch-pavan.github.io/mla-networth/	npm run build:pages

Pushes to main deploy both targets (Vercel via Git integration; Pages via .github/workflows/pages.yml).

Both builds exclude unfinished Cloudflare D1 routes under app/api/. The public UI reads public/data/. Canonical metadata uses NEXT_PUBLIC_SITE_URL when set, otherwise the host’s production domain.

Data refresh runbook

Read the data contract before refreshing. Preserve source attribution and review generated diffs.

Sitting MLAs from a reviewed ADR PDF (needs Python + pdfplumber):

python3 scripts/extract-adr-report.py <path-to-reviewed-report.pdf> public/data/adr-sitting-mlas-2025.json

Candidate and winner archives load the reviewed election manifest. Run candidates before winners:

npm run data:candidates
npm run data:winners
npm run data:history
npm run verify

These hit third-party sites and can produce large diffs. Check record counts, coverage flags, timestamps, hashes, money-status totals, by-election totals, spot samples, and test failures before committing. Generators decode only the recognized MyNeta packed-row format; unknown scripts and non-contiguous ranks fail instead of silently dropping rows.

Candidate amounts use parsed, nil, masked, and missing. Only nil means explicitly no amount; masked/missing stay null. Profile lookups write to ignored work/myneta-profile-cache/. Winner amounts conflicting with candidate profiles land in moneyConflicts.

Optional SQL seed (does not migrate or touch a remote DB):

npm run data:import -- <normalized.csv-or-json> <output-seed.sql>

Never commit source PDFs/credentials without review. D1 is not part of production deploys.

Scripts

Command	Purpose
npm run dev	Local development
npm run build	vinext production build
npm run build:pages	Static export for GitHub Pages
npm run build:vercel	Next.js production build for Vercel
npm test	Build + dataset/UI smoke tests
npm run lint	ESLint
npm run typecheck	Typecheck
npm run verify	Lint, typecheck, build, tests
npm run data:candidates	Refresh candidate archive
npm run data:winners	Refresh winner archive
npm run data:history	Refresh recontest history
npm run data:loksabha	Refresh Lok Sabha candidates, winners, sitting MPs
npm run data:rajyasabha	Extract Rajya Sabha sitting MPs from ADR PDF
npm run data:match-geo	Match assembly constituencies to AC geo
npm run data:match-pc-geo	Match Lok Sabha constituencies to PC geo
npm run data:import	Import affidavits into SQL seed
npm run db:generate	Generate Drizzle migrations

See data/README.md for the data contract and source rules.