An interactive financial disclosure and conflict-of-interest tracker for the executive branch.
Congress has well-known stock trackers like Capitol Trades and Quiver Quantitative. Executive branch officials file under the same STOCK Act rules — 30 to 45 days per trade — but their filings get far less scrutiny. Open Cabinet turns those filings into searchable timelines, compliance flags and company-level lookups.
Live: open-cabinet.org
- Tracks financial transactions filed by cabinet secretaries, agency heads and senior government officials
- Parses OGE filing PDFs into searchable, sortable data with D3 visualizations
- Flags late filings, tracks compliance and surfaces potential conflicts of interest
| Metric | Value |
|---|---|
| Officials tracked | 39 |
| Transactions | 10,445 |
| Estimated value | ~$4.3B |
| Late filings | 7,050 |
| Companies searchable | 430 |
| News articles linked | 35 |
| Source filing PDFs linked | 189 |
Every number in this table is checked against public/data/full-dataset.json by an automated test (lib/readme-stats.test.ts). CI fails if the table drifts from the published dataset.
| Page | Route | Description |
|---|---|---|
| Directory | / |
All officials with transaction counts, sortable, expandable |
| All Trades | /all |
Swim lane chart — every transaction on one canvas with time filters |
| Companies | /companies |
Search by ticker, see which officials traded each stock |
| Late Filings | /late-filings |
Accountability page: who files late and what the law says |
| Overview | /dashboard |
Aggregate buy/sell ratio, asset treemap, official rankings |
| Official Detail | /officials/[slug] |
Transaction timeline, trade table, news coverage |
| Company Detail | /companies/[ticker] |
Who in government trades this stock |
| About | /about |
Methodology, legal basis, AI transparency, feedback form |
| Download | /download |
CSV and JSON exports of the full dataset |
| Admin | /admin |
Pipeline status, review queue, data validation (auth-gated) |
All data comes from the U.S. Office of Government Ethics. Transaction reports (278-T Periodic Transaction Reports) are filed under the STOCK Act and the Ethics in Government Act (5 U.S.C. Section 13107). Federal government documents carry no copyright (17 U.S.C. Section 105).
OGE API ──▶ scripts/ingest-new-filings.ts ──▶ data/officials/*.json (source of truth)
│
├─▶ scripts/rebuild-index.ts ──▶ data/meta/officials-index.json
└─▶ scripts/generate-exports.ts ──▶ public/data/*.json, *.csv
Next.js App Router reads data/ at build time ──▶ static pages (650+ prerendered)
Neon PostgreSQL + Better Auth ──▶ /admin panel, email alerts (Resend)
The public site is served entirely from static JSON committed to this repo — no live database reads on public pages. The database backs the admin panel, pipeline run history and the email alert system.
Open Cabinet uses two scheduled paths:
- Monitor — Vercel Cron polls the OGE API weekly, diffs exact 278-T PDF URLs against tracked source filings, records the run and emails the result.
- Ingest — GitHub Actions runs the static JSON ingest weekly or on demand, downloads new PDFs, parses them with Claude, validates the data, regenerates exports and opens a PR for review.
The ingest path runs these stages:
- Extract — natural-pdf (by Jonathan Soma) extracts text page by page.
- Parse — Claude Opus/Sonnet structures each page into transaction data.
- Validate — Automated checks: schema, tickers, regression tests, anomaly detection.
- Store — Neon PostgreSQL with UNIQUE deduplication and batchId for rollback.
- Alert — Email notifications for errors, credit exhaustion or new filings via Resend.
pnpm run pipeline # Full run: check, download, parse, validate, insert
pnpm run pipeline -- --dry-run # Check only, no inserts
pnpm run pipeline -- --verify # Parse then cross-check with OpenAI
pnpm run ingest-filings # Update static JSON from new OGE PDF URLs
pnpm run check-filings -- --dry-run # URL-diff OGE without writing state
pnpm run validate # Run validation suite against data
pnpm run parse-pdf <file> # Parse a single PDF
pnpm run check-news # News coverage search guidance
pnpm run seed # Seed database from JSON files| Model | Provider | Cost/PDF | Role |
|---|---|---|---|
| Claude Sonnet 4.6 | Anthropic | ~$0.02 | Default parser |
| Claude Haiku 4.5 | Anthropic | ~$0.01 | Budget option |
| Claude Opus 4.6 | Anthropic | ~$0.06 | Verification |
| GPT-5.4-mini | OpenAI | ~$0.01 | Cross-provider check |
| GPT-5.4-nano | OpenAI | ~$0.003 | Cheapest fallback |
- Next.js 16 (App Router, static generation, 650+ pages prerendered)
- React 19 + TypeScript
- D3.js v7 sub-modules for all visualizations
- Tailwind CSS 4
- Neon PostgreSQL (serverless) + Drizzle ORM
- Better Auth with Google OAuth (admin panel)
- Anthropic SDK + OpenAI SDK for PDF parsing
- Resend for email notifications
- Vercel (Pro) for hosting and lightweight cron monitoring
- GitHub Actions for weekly pipeline ingest and PR creation
- pnpm for package management
git clone https://github.com/tbrown034/open-cabinet.git
cd open-cabinet
pnpm install
cp .env.example .env.local # Fill in your API keys
pnpm dev # http://localhost:3003See .env.example for the full list. Required:
ANTHROPIC_API_KEY— Claude API for PDF parsingOPENAI_API_KEY— Cross-provider verification (optional)DATABASE_URL/DATABASE_URL_UNPOOLED— Neon PostgreSQLBETTER_AUTH_SECRET— Session signingGOOGLE_CLIENT_ID/GOOGLE_CLIENT_SECRET— Admin OAuthRESEND_API_KEY— Email notificationsCRON_SECRET— Vercel Cron authentication
The research/ directory contains six internally sourced research briefs (180+ pages) covering:
- STOCK Act and federal ethics law
- Case law and legal precedent
- News coverage of executive branch financial conflicts
- OGE structure and data landscape
- Late filing patterns and enforcement
- The divestiture process
All briefs follow SPJ Code of Ethics standards with inline citations. See research/README.md for the index.
pnpm test # Vitest unit tests (digest scoping, alert tokens,
# office-line formatting, README stats vs dataset)
pnpm lint # ESLint (app and lib; scripts/ excluded by design)
pnpm typecheck # tsc --noEmit across app, lib and scriptsGitHub Actions runs all three on every push and pull request (.github/workflows/ci.yml). A second workflow (oge-pipeline.yml) runs the weekly OGE ingest and opens a data PR when new filings appear.
pnpm run validate # Schema + golden file regression tests
/copy-review # AP style + journalism ethics audit (Claude Code command)
/anomaly-check # Data quality + contextual anomaly detectionThis tool aggregates public records. The Ethics in Government Act's news media exception explicitly permits dissemination of financial disclosures to the general public. Multiple for-profit companies (Capitol Trades, Quiver Quantitative, Unusual Whales) operate similarly with congressional data. No enforcement action has ever been brought against a disclosure aggregator.
For informational and journalism purposes only. Not investment advice.
- PDF parsing: Claude Sonnet (default) with OpenAI cross-verification
- Official summaries: AI-generated from transaction data, reviewed for accuracy
- News coverage: AI-assisted search, all linked articles are real published pieces
- Codebase: Built by Trevor Brown with the assistance of Claude Code
- What AI does NOT do: No fabricated data, no editorial judgments, no decisions about who to track
See the About page for full AI transparency disclosure.
Found a data error? Open an issue or use the feedback form on the site.
Trevor Brown — investigative data journalist turned web developer. 15 years of political reporting including six years covering elections, dark money, financial disclosures and government accountability at Oklahoma Watch. Built Oklahoma's first statewide financial disclosure database.