cityscope 0.4.0

Find your next real estate market in minutes, not months.

A pip-installable Python package that pulls public government data about every major US metro and city — population growth, job growth, wages, unemployment — and stores it locally in SQLite for analysis. Use it as a library, from the CLI, or build your own UI on top.

Install

pip install cityscope

Or with uv:

uv add cityscope

Python API

from cityscope import api

# Fetch data from public sources
api.fetch("census_population")
api.fetch("bls_employment", skip_laus=True)

# Query as a DataFrame
df = api.to_dataframe(metric="population_change_pct", geo_type="metro", year=2024)
print(df.sort_values("value", ascending=False).head(10))

# Or as a list of dicts
rows = api.query(metric="employment_change_pct", geo_type="metro", year=2024, limit=10)
for row in rows:
    print(f"{row['name']}: {row['value']:+.1f}% job growth")

# Look up stats for a specific address
report = api.lookup("1600 Amphitheatre Pkwy, Mountain View, CA", auto_fetch=True)
print(f"Metro: {report.metro.name} — {report.metro.metrics}")
print(f"City:  {report.city.name} — population {report.city.population:,}")
print(f"County: {report.county.name}")

# Configure (optional — works with defaults)
api.configure(db_path="my_data.db", min_population=100_000)

# Check what you have
api.status()
api.list_sources()
api.get_geographies(geo_type="metro", min_population=500_000)

CLI

# Fetch data
cityscope fetch census_population
cityscope fetch bls_employment --skip-laus
cityscope fetch --all

# Query
cityscope query -m population_change_pct -g metro -y 2024
cityscope query -m employment_change_pct -g metro -y 2024 -n 10
cityscope query -m avg_annual_pay -y 2024 -n 15

# Look up stats for an address
cityscope lookup "1600 Amphitheatre Pkwy, Mountain View, CA"
cityscope lookup "123 Main St, Austin, TX" --auto-fetch

# Info
cityscope sources
cityscope status

Commands

Address Lookup

The lookup command geocodes a US address (via the free Census Geocoder) and returns stats for the enclosing metro, city, and county:

cityscope lookup "1600 Amphitheatre Pkwy, Mountain View, CA" --auto-fetch

With --auto-fetch, cityscope will fall back to fetching missing data from source APIs on-the-fly (useful for addresses in counties or smaller cities not covered by the default 200k+ metro/city fetch). Results are cached in the local DB.

Global flags: -v (verbose logging), -c PATH (custom config file).

Fetch Flags

What Data You Get

370+ metros and cities (200k+ population), each tracked across 9 metrics over 5 years (2020–2024):

All data is free, public domain, pulled directly from federal APIs.

Configuration

Optional — everything works with defaults. For higher API limits:

# config/settings.yaml
census:
  api_key: null    # Free: https://api.census.gov/data/key_signup.html
bls:
  api_key: null    # Free: https://data.bls.gov/registrationEngine/
storage:
  db_path: data/cityscope.db
pipeline:
  min_population: 200000

Or configure programmatically:

api.configure(census_api_key="your_key", bls_api_key="your_key")

Architecture

Census API ──┐                     ┌── Python API (cityscope.api)
             ├── Pipeline ── SQLite DB ──┤
BLS QCEW  ──┘    (fetch)          └── CLI (cityscope)

Data flows: Source → Pipeline → SQLite → API/CLI/your code.

Adding Data Sources

Each source is a self-contained class with @SourceRegistry.register:

from cityscope.core.registry import SourceRegistry
from cityscope.core.source import DataSource
from cityscope.core.models import FetchResult

@SourceRegistry.register
class MySource(DataSource):
    source_id = "my_source"
    name = "My Data Source"
    description = "What it provides"

    def fetch(self, **kwargs) -> FetchResult:
        ...

Add the import to src/cityscope/sources/__init__.py — it auto-registers in CLI, API, and pipeline.

Dashboard

For a visual dashboard, see urban-research-ui.

Roadmap

• Rent data (HUD Fair Market Rents, Zillow ZORI)
• Home price index (FHFA HPI)
• Crime stats (FBI Crime Data Explorer)
• School quality (NCES)
• Walkability (EPA Smart Location Database)
• Migration flows (IRS SOI county-to-county)
• Neighborhood-level data (Census tract)
• Composite scoring engine

See data_sources.md for research on 50+ public data sources.

Contributing

Pull requests welcome. The easiest way to contribute is adding a new data source — the plugin architecture makes it straightforward.

License

MIT

Built with Claude Code (Claude Opus 4.6).