jobgauge

A static U.S. labor-market dashboard.

Live dashboard: https://smkwray.github.io/jobgauge/

jobgauge tracks the U.S. labor market using curated series from official statistical agencies (BLS, Census, BEA, DOL), with some series mirrored via FRED for context. A Python pipeline ingests and normalizes the data and exports a static data contract (JSON); a zero-build single-page app renders it on GitHub Pages with no server at request time.

The dashboard lets you explore indicators, build comparisons, derive new series from a formula, rank states on a map, and export charts to PNG, SVG, or CSV with full source provenance. Every view is shareable through its URL.

How it works

catalog/  (indicator + source metadata, YAML)
   │   Python pipeline: fetch → normalize → transform → export
   ▼
site/public/data/    catalog.json · manifest.json · latest.json · series/<id>.json
site/public/search/  index.json (client-side fuzzy search)
   │
   └─ rendered by the static single-page app in site/public/  →  GitHub Pages

Each observation in the series files carries precomputed transform columns (changes, year-over-year, rolling averages, an index), so the dashboard reads consistent values without recomputing. The contract is documented in docs/STATIC_DATA_CONTRACT.md.

Data sources & attribution

Data comes from the U.S. Bureau of Labor Statistics, Census Bureau (QWI / LEHD), Bureau of Economic Analysis, and Department of Labor, with some series mirrored via FRED for convenience. Source owner, series ID, units, frequency, seasonal adjustment, and geography are preserved throughout. See NOTICE.md.

Local development

Serve the dashboard against the bundled data:

cd site/public
python3 -m http.server 8137 --bind 127.0.0.1
# open http://127.0.0.1:8137/

Run the data pipeline (free API keys are required only for live refreshes):

python -m venv .venv && source .venv/bin/activate
pip install -e ".[dev]"
cp .env.example .env          # add keys
make validate-catalog
make export-static && make build-search

See docs/REFRESH_AUTOMATION.md for refresh and export targets.

Repository layout

• site/public/ — the static dashboard (HTML/CSS/ES-module JS + ECharts) and the exported data contract
• src/labor_dashboard/ — the Python data pipeline (providers, transforms, static export)
• catalog/ — indicator and source metadata (YAML)
• tools/ — a standalone command-line tool for querying the exported data offline
• docs/ — data contract, tool reference, and refresh notes
• .github/workflows/ — CI, scheduled data refresh, and Pages deploy

License

Released under the MIT License. The underlying data originates from U.S. public-sector statistical agencies; see NOTICE.md for source attribution requirements.