Clearance Critter

Fetch one current deal by the id returned from search_deals, with its full price provenance, baseline, expiry and purchase link. Returns an error if that listing is no longer current — it never returns a lapsed offer. Read-only.

get_deal

Fetch one tracked product by id, with its last observed price and time, stock reading, and — only if it is a current deal right now — the full deal. Returns an error for an id that is not tracked. Read-only.

get_product

Return what Clearance Critter has actually published about a listing's price history: the aggregate baseline computed from its own observations (price, method, observation count, distinct days, span) and the retailer's own comparison price when there is one. Clearance Critter does not publish a per-day price series, so no time series is returned and none is inferred; a listing with no published baseline is reported as having none. Read-only.

get_price_history

List Clearance Critter's departments and categories with how many current deals each one holds, plus the merchants currently represented. A department with zero deals is reported as zero rather than hidden. Read-only.

list_categories

Search current Clearance Critter deals by keyword, department, category, merchant, price ceiling, condition, minimum advertised discount, or the deterministic 0-100 `quality` score this server computes from its own observed price history. Every filter and the sort are applied to the whole current catalog before the page is cut, so a page of `sort: "quality"` results is the best in the catalog, not the best of an arbitrary slice. The default sort stays "recommended" (our own ranking, with a small disclosed nudge for Amazon listings); it is not a recency order. Examples. Best ten deals right now: {"sort":"quality","minQualityScore":60,"limit":10}. Best in one department under a budget: {"department":"Office","maxPrice":200,"sort":"quality"}. Biggest dollar savings with real history behind them: {"sort":"savings","minBaselineDays":7,"referenceKind":"observed-baseline"}. Then call get_price_history with an id from the results to see the baseline that score was measured against. Returns only offers that are still current, with the provenance of every price and an affiliate disclosure on Amazon links. Unknown arguments are rejected rather than ignored. Read-only; it cannot buy anything.

search_deals

Search every product Clearance Critter is tracking — deals and non-deals — by keyword, category, department, merchant, budget and stock. Words match whole words, never fragments, and a result must contain every word of a one- or two-word query (all but one in three of a longer one); no match returns an empty list, not the nearest thing. Default order is relevance: products matching more of your words first, then current deals, then products with a recent in-stock reading, then a recent price. A deal that matches your words worse never outranks one that matches better. Prices are last observations with their time, never live prices. `priceFresh` is true only for a reading under 24 hours old. With `minPrice`/`maxPrice`, a price counts as fitting the budget by default when it is fresh, or when it is a `recentProviderReport`: a price data provider's price we retrieved under 24 hours ago whose original observation time the provider did not give. Relay those as provider-reported at the retrieval time with unknown original age, and tell the shopper to confirm at the retailer; `providerReportsInBudget` counts them. A dated reading over 24 hours old, an older retrieval, or an undated feed price is left out unless `includeStalePrices` is true; the answer says how many were left out. Stale products are otherwise still returned, with their age. Only products with `isDeal: true` are deals. Read-only; unknown arguments are rejected.

search_products