尚未收集仓库信息。
尚未检测到更新。GitHub 发布版本后会显示在这里。
Municipal Financial Risk Score — open screening of local government Single Audits
翻译准备中
正在加载完整 README。
查看完整 README ↗Open screening of local government Single Audits · live at civicauditlens.org
Free and public. Covers all 50 states, DC and the U.S. territories.
Local governments that spend federal awards above the Single Audit threshold must file an audit with the Federal Audit Clearinghouse (FAC). The GAO found that the FAC cannot identify entities that should file but do not (GAO-24-106173), and the findings it does hold are hard for oversight bodies to use at scale.
This project:
/gov/{STATE}/{UEI}/{name}) that can be shared, cited and found in search,
plus a printable brief for each state, a batch lookup for pass-through entities, and a weekly "What changed" page
with RSS feeds./#compare/IL:UEI1,IL:UEI2): score and its four parts, peer
percentile, findings, filing timeliness, warning signs, programs, pass-through entities and score history since 2016,
with a shareable link, CSV download and suggestions of similar-size governments in the same state./#insights-auditors-IL-sikich): the governments whose
latest Single Audit a firm signed, with each government's score. A lookup only: it does not score, rate or rank firms.Scores are screening indicators, not audit opinions and not findings of fraud or wrongdoing.
| Component | What it does | Where |
|---|---|---|
| Municipal Financial Risk Score (national) | Scores every U.S. local government Single Audit from FAC data and flags apparent stopped filers; public website | pipeline/, site/ |
| Illinois fiscal pilot | Fiscal-stress screening from financial-statement ratios (reserves, operating margin, debt service) for Illinois municipalities | illinois-fiscal-pilot/ |
Both share the same aim: open, explainable methods that help oversight bodies find local governments that may need a closer look, earlier.
FAC API (api.fac.gov) ──weekly GitHub Action──▶ pipeline/build_dataset.py ──▶ site/data/*.json, *.csv
│
Cloudflare Pages (free) serves site/ ◀┘ → your domain
pipeline/scoring.py: the scoring rules (all point values in one place)pipeline/build_dataset.py: downloads FAC data, scores it, detects stopped filers, writes the data filespipeline/analytics.py: the additional analyses (requirements, programs, pass-through, timeliness, persistence,
themes, peers, weekly changes, RSS, sitemap). Each one is optional: if it fails, the scores still publish and the
site says that analysis is unavailable that weekpipeline/test_scoring.py, pipeline/test_analytics.py: tests that lock in the rulessrc/worker.js, src/gov.js: the Cloudflare Worker (feedback form, one page per government, sitemap and feeds);
tests/gov.test.mjs tests the government pages (node --test tests/gov.test.mjs)site/validation.json: results of manual spot checks of the stopped-filer list (edit by hand; shown on the
Data quality page)pipeline/make_sample_data.py: fictional data for previews (shows a SAMPLE banner)site/: the web app (plain HTML/CSS/JS, no build step)Run it locally (from the repository root):
pip install pytest && (cd pipeline && python -m pytest -q)
FAC_API_KEY=your_key python pipeline/build_dataset.py # real data (history from 2016; ~20-30 minutes)
python -m http.server -d site 8000 # open http://localhost:8000
Setup steps for hosting, the API key, your domain and a DOI are in SETUP.md.
The methodology is on the app's Methodology page and in pipeline/scoring.py.
Use the Give feedback button in the app, or open an issue. Auditors, oversight agencies, pass-through entities, finance officers and researchers are especially welcome.
Code: MIT. Data: derived from FAC public-domain data and released under CC0.