Top level

foundboolean

Whether a catalog product confidently matched. false = limited-coverage shape.

product_namestring

Canonical catalog name of the matched product.

brandstring

Manufacturer or brand name.

headlinestring

One-line editorial identity for the product.

summarystring

Long-form synthesized verdict (dropped in compact mode).

buyary_scorenumber

The composite 0–100 score.

source_countnumber

How many independent sources were synthesized.

as_ofstring

ISO timestamp of the evaluation snapshot.

Scores

dimension_scores[]object[]

dimension (experts / users / value / …), score, and a synthesized summary.

feature_scores[]object[]

All 27 criteria across 7 dimensions, each with a name and 0–100 score.

Claims

highlights[]object[]

Verified strengths: title + detail, each traceable to sources.

considerations[]object[]

Reported issues, same shape — the honest half of the synthesis.

Consensus & commerce

source_consensus[]object[]

Per-source rows: source, source_type, score, stance, and a per-source summary.

offers[]object[]

Live offers: retailer_name, price, price_as_of, and a monetizable buy_url.

starting_price_usdnumber?

List price at launch, when known.

buy_atstring?

Primary purchase link.

Resolution

resolutionobject

Match confidence plus other_matches[] (name + match_score) so your agent can sanity-check the match.

provenanceobject

Where the synthesis came from: catalog snapshot and evaluation lineage.