Public Filing System · Congressional Disclosure Data

Documentation

Quick start

Install

Python 3.10+. No database server. DuckDB ships as a library dependency.

git clone https://github.com/DMulajkar/Quantgress.git
cd Quantgress
py -m pip install -r requirements.txt

Five steps

# 1. Validate every parser offline (instant, no network)
py scrape_senate.py --selftest
py scrape_house.py --selftest

# 2. Bounded live runs to confirm access before a full backfill
py scrape_senate.py --limit 20
py scrape_house.py --year 2026 --limit 25

# 3. Resolve tickers embedded in free-text asset names
py entities.py --dry     # preview, writes nothing
py entities.py           # write recovered tickers

# 4. Query
py q.py "SELECT * FROM trades LIMIT 10"

# 5. Serve it
py -m uvicorn api:app --reload
curl http://127.0.0.1:8000/trades?tkr=NVDA&limit=5

A full Senate + House historical backfill is ~1.3 hours against 2,411 filings at a self-imposed 2-second rate limit. Run --limit N before an unbounded pass.

Prev: Docs overview Next: Datasets