elgarde-inspection

A structured pre-purchase used-car inspection as data plus a small pure-Python implementation: a multilingual checklist, curated model-specific known defects, and the 1–5 scoring rules that turn answers into one verdict.

No dependencies, no network calls, no configuration. This is the same canon that powers the Elgarde app and its free API, and it passes the same conformance vectors as every other port.

pip install elgarde-inspection

Use

from elgarde_inspection import Car, assemble, all_items, localize, overall, progress

groups = assemble(Car(make="Volkswagen", model="Golf", fuel_type="diesel"))
[g["id"] for g in groups]
# ['ext', 'int', 'eng', 'road', 'docs', 'fuel.diesel', 'defects']

len(all_items(groups))
# 33  — 25 base items, 3 diesel-specific, 5 known VAG weak points

The checklist adapts to the car: a diesel gets DPF and glow-plug items, an EV gets state-of-health and charging items, and a recognised model gets its documented weak points — the DQ200 dry clutch, the PureTech wet belt, the N47 timing chain.

Read it in one language

for group in localize(groups, "pt"):
    print(group["title"])
    for child in group.get("children", []):
        for item in child.get("items", []):
            print("  ", item["prompt"], "(critical)" if item["critical"] else "")

Content is authored in English, Portuguese, Russian, Ukrainian and French; defect text in English and Portuguese, falling back to English.

Score an inspection

Answers are 1–5, or -1 for not applicable and -2 for not inspected. Omit an item to leave it unanswered.

scores = {item["id"]: 5 for item in all_items(groups)}
scores["docs.all.mileage"] = 1          # odometer does not match the records

overall(groups, scores)                 # 1.0
progress(groups, scores)                # (33, 33)

The average was 4.86. A critical item scoring at or below 2 clamps the verdict down to that score, because a car with good averages and a failed safety or provenance check is not a good car. That clamp only ever lowers the verdict, and an unanswered critical item has no effect.

Group rollups are average, weighted (by item["weight"]) or worst_case, per group. The full algorithm is written down in spec/SCORING.md.

Just the known defects

from elgarde_inspection import match_defect_rule

rule = match_defect_rule("Peugeot", "208")
[(d["id"], d.get("critical", False)) for d in rule["defects"]]
# [('puretech.wetbelt', True), ('ep6.chain', False), ('hdi.turbo', False)]

None means nothing is curated for that model yet — not that the model is trouble-free.

Honest limits

Elgarde is a self-service checklist. It is not a certified, mechanical or legal inspection, not professional advice, and not a vehicle-history or title service. A defect entry describes a known weak point on a model — something worth checking — and is never a claim about any individual car.

Licence

Code MIT; the inspection data is CC-BY-4.0 — attribution to https://elgarde.com/ is the only condition.

Source, the canon, and the conformance suite: https://github.com/ElgardeCom/elgarde