us-home-safety-score

Compute residential Home Safety Scores (0-100, A-F) from U.S. federal data. Zero dependencies. Pure Python 3.9+.

This is the Python port of the us-home-safety-score npm package. Both implementations produce identical scores for the same inputs (verified bit-for-bit across a 640-case grid, including rounding edge cases).

The Home Safety Score is a composite index that evaluates environmental risk for any U.S. ZIP code across four dimensions:

• Water quality — EPA SDWIS violation history
• Lead exposure — EPA Lead and Copper Rule sampling
• Radon risk — EPA county-level radon zones
• Flood risk — FEMA NFIP historical claims

Higher score = safer. It is the same algorithm used on ZipCheckup.com to score tens of thousands of ZIP codes.

Install

pip install us-home-safety-score

Quick Start

from us_home_safety_score import compute_safety_score

result = compute_safety_score({
    "totalViolations": 3,
    "healthViolations": 1,
    "leadLevel": 0.008,     # mg/L (90th percentile)
    "radonZone": 2,         # EPA zone: 1=High, 2=Moderate, 3=Low
    "floodClaims": 25,      # FEMA NFIP historical claims
})

print(result)
# {
#   'score': 67,
#   'grade': 'C',
#   'components': {
#     'water': {'score': 21, 'maxPoints': 25, 'weight': 0.25},
#     'lead':  {'score': 18, 'maxPoints': 25, 'weight': 0.25},
#     'radon': {'score': 13, 'maxPoints': 25, 'weight': 0.25},
#     'flood': {'score': 15, 'maxPoints': 25, 'weight': 0.25},
#   },
#   'componentCount': 4,
# }

Field names use camelCase (totalViolations, leadLevel, floodClaims, …) to stay 1:1 with the npm package's input contract.

Scoring

When FEMA flood data is unavailable, the score falls back to a 3-component mode (each component worth 33 points instead of 25). Missing lead or radon data uses a neutral assumption (2/3 of max points) rather than penalizing or rewarding the location.

Grade thresholds: A ≥ 85, B ≥ 70, C ≥ 55, D ≥ 40, F < 40.

Full methodology: https://zipcheckup.com/about/home-safety-score/

API

from us_home_safety_score import compute_lead_risk

compute_lead_risk({"leadLevel": 0.020})
# {'risk': 'high', 'probability': 0.85, 'exceedsActionLevel': True,
#  'description': 'Lead level exceeds the EPA action level (15 ppb). ...'}

Data sources

• EPA SDWIS — Safe Drinking Water Information System (violations, Lead and Copper Rule sampling)
• EPA Radon Zones — county-level radon potential map
• FEMA NFIP — National Flood Insurance Program claims

This package contains only the scoring logic — you supply the federal data inputs. Pre-joined per-ZIP data is available via the us-water-quality-data package and the ZipCheckup API.

License

MIT © ZipCheckup