- Brand
- Unknown
- Category
- Pending
- Primary Subcategory
- Pending
Integration details
Description
Your Amazon business has a story. Ask sellerfox. Connect your sellerfox account to ChatGPT and turn your Amazon data into answers you can act on. Explore what drives your sales, which products contribute most to your profit and how your advertising is performing — simply by asking. Compare periods, explore cost breakdowns and connect the dots between sales, traffic, search performance and keyword rankings. Dig deeper with follow-up questions to understand what’s working, what’s changing and what deserves your attention. Requires a sellerfox account with connected Amazon data.
- Integration type
- Plugin
- Verification status
- Not applicable
- Platform
- ChatGPT
- Category
- Pending
- Primary Subcategory
- Pending
- Secondary Subcategories
- None listed
- Brand
- Unknown
- Access
- Account required
- First tracked
- 2026-10-15
- Tool count
- 22
- Geography
- US
The broad Category that contains the Primary Subcategory.
The Primary Subcategory used for this profile’s headline score.
Other Subcategories where the Integration is listed.
Your score is coming
ChatGPT now suggests Plugins on its own when they match a user's request.Your Plugin Discovery Score measures how often yours appears, and it will show here as soon as it’s ready.
What discovery looks like

Get alerts for sellerfox
Get updates when sellerfox’s Discoverability Score or category rank changes.
Competitive lineup
22 tools agents can invoke
Returns a paginated Best Seller Rank (BSR) snapshot per (asin, marketplace), with each row carrying the latest main + sub-1 category rank, the rank change against about 7 days earlier (delta7d), and 30-day revenue. Sortable by `mainRank`, `sub1Rank`, `delta7d`, or `revenue`. Supports parent-family grouping with best / average / worst aggregation across child variations. USE THIS FOR: "show me my top BSR products on amazon.de", "which products dropped in rank this week" (sortBy "delta7d" descending), "current main-category rank for these ASINs" (asinMarketplaces[] filter), "best parent product families by BSR" (groupBy "parent"), "worst rank in each product family last update" (groupBy "parent" + aggregation "worst"). The default is a catalog-wide ranking across every enabled marketplace. IMPORTANT: This tool returns a CURRENT-STATE snapshot — there is no date range. For BSR time-series and 30/90/180/365-day rank deltas of a single product, use get_bsr_trends instead. get_bsr_trends is the right tool for "chart the BSR history of B07XYZ1234 over the last 90 days" or any other historical question. get_bsr_ranks vs get_bsr_trends: snapshot vs trend; many products vs one product. Returns: { data: [...], meta: {...} } — identical shape to the REST /statistics/bsr/overview endpoint, wrapped in the MCP text-content envelope. Each row carries asin, marketplaceId, mainRank, mainTitle, sub1Rank, sub1Title, revenue, prevMainRank, delta7d, lastUpdatedAt, plus the joined product metadata (productId, title, imageSmallUrl, isParent, parentAsin, skus). prevMainRank is the product's main rank on its marketplace's comparison day: the newest day 7 to 11 days before the snapshot date on which any active product has a rank (exactly 7 days back when that day has data, up to 11 days back when days are missing). delta7d is mainRank minus prevMainRank. Both are null when the product has no rank on that comparison day; null means unknown, not unchanged, so never report it as "no change". Rows with a null delta7d sort after all known values in both directions. When groupBy is "parent", rows also include childCount. Meta includes pagination, sort context, and snapshotDates per requested marketplace (ISO code and date, or null when no positive BSR snapshot exists in the last 30 days). Snapshot dates older than 3 days are stale; say so when you report them. Marketplaces can have different snapshot dates; do not describe older snapshots as live ranks.
Returns the daily Best Seller Rank (BSR) time series and 30 / 90 / 180 / 365-day rank deltas for a single ASIN on a single marketplace over a date range. USE THIS FOR: "chart the BSR history of B07XYZ1234 on amazon.de over the last 90 days", "how has my main-category rank moved this quarter for ASIN B0ABCDEF", "what is the 30-day BSR delta for B0ABCDEF12 on amazon.com", "graph the daily rank trend including sub-category placements". IMPORTANT: This tool is scoped to ONE (asin, marketplaceCode) pair per call. For a current rank SNAPSHOT across the whole catalog (paginated, sortable, with parent/child grouping) use get_bsr_ranks instead — get_bsr_ranks is the right tool for "which of my products dropped in rank this week", "show me my top BSR products", or any "current state" question. Use get_bsr_trends for time-series and delta questions about a single product. The ASIN must belong to an existing, not deleted product of this workspace on that marketplace (get_products lists them); otherwise the tool returns an error. Date ranges are capped at 730 days. Days with no rank data are omitted from the series; sub-category ranks may be null on rows where the main rank is present. Returns: { timeSeries: [...], deltas: {...} } as JSON text and structured data. Each timeSeries row carries date plus mainRank/mainTitle/mainLink and sub1/sub2/sub3 rank, title, and link. Deltas are computed as (latestRank - rankNDaysAgo) using nearest-day matching within a +/- 2 day window; positive values mean the rank got worse (higher number), negative values mean it improved. Deltas may be absent when the lookback window has no matching data.
Returns aggregated ranking and placement KPIs for tracked keywords across the workspace, with optional period-over-period comparison. USE THIS FOR: "how are my keywords ranking", "best and worst keyword positions this week", "keyword ranking change vs last month", "which keywords have the most placements on DE". IMPORTANT: Always pair fromComparison and toComparison for trend questions, otherwise the response has no comparison block. Filter to specific products with `asinMarketplaces[]` tuples; omit to query every tracked keyword in the workspace. For ranking time series use get_keyword_trends instead. Ranking KPIs are reversed-scale (lower position is better) — the server handles direction automatically when sorting. Date ranges are capped at 730 days. Returns: One row per keyword (keyword, marketplaceCode) with KPI values. When a comparison range is supplied, `comparison` follows `data` row for row, so `comparison[i]` is always the same keyword as `data[i]`; each row also carries its own keyword and marketplaceCode if you prefer to join by key, and a keyword with no data in the comparison window appears with null KPI values. Response uses the English KPI label (e.g. "Organic Ranking") as the JSON key, not the code; pass codes from get_kpi_definitions in input fields.
Returns time-series ranking KPIs for tracked keywords, aggregated by day, week, or month. USE THIS FOR: "chart my keyword position over the last 30 days", "show me the rank trend for keyword X", "how have my organic placements moved this quarter", "graph rank history of the keywords tracked for the product family of ASIN B07XYZ". IMPORTANT: Pick one KPI per call — time-series tools return a single metric per series. Filter to specific products with `asinMarketplaces[]` tuples; omit to return all tracked keywords. With a product filter, each rank is the average across the whole product family (parent and all child ASINs); there is no per-child ranking. For aggregated totals with comparison use get_keyword_kpis instead. Use granularity DAY for short ranges, WEEK for quarters, MONTH for longer spans. Ranking KPIs are reversed-scale (lower position is better). Date ranges are capped at 730 days. Returns: One row per keyword with a timeSeries array aligned to meta.headerDates. Response meta.kpi carries the English label (e.g. "Organic Ranking"), not the code; pass the code from get_kpi_definitions in the `kpi` input field.
Returns the list of keywords tracked in the workspace, with marketplace and favorite status. USE THIS FOR: "what keywords am I tracking", "list my tracked keywords for DE", "show favorite keywords", "do I track <some phrase>". IMPORTANT: This is the keyword catalog, not ranking data. For ranking metrics call get_keyword_kpis or get_keyword_trends. For untracked search terms Amazon already attributes traffic to, use get_sqpr_kpis. Soft-deleted keywords are excluded; labels are stored lowercase and searchText is matched case-insensitively. Returns: Keyword list with id, label, marketplaceCode, and isFavorite. Filterable by marketplace, favoritesOnly, and searchText.
Brand-level SQPR is disabled because report data is unavailable. Brand SQPR KPI codes cannot be requested; ASIN SQPR and general search-query metrics remain available. Returns the catalog of KPI codes available across product, marketplace, keyword, SQPR, and profit tools — each with English label, unit, category, reversed-scale flag, and B2B-variant linkage. USE THIS FOR: "what does ACOS mean", "which KPI is conversion rate", "explain how this metric is calculated", interpreting any KPI code returned by other tools. IMPORTANT: This is reference metadata, not workspace state. Codes and labels are stable across workspaces; labels are always English. Call once per session; the catalog does not change between requests. KPI-taking tools accept these `code` values exactly, so do not translate the label back into a code. Check `domains` before passing a code: it is only valid on a tool whose entity appears there (product, marketplace, keyword, searchQuery, profit), and passing one outside its domains is rejected with a validation error. A `isReversedScale: true` flag means a SMALLER value is the better outcome (ACOS, refund rate, cost-per-*). Returns: One entry per KPI code with code, label, unit, category, domains, isReversedScale, and optional b2bVariantOf. Categories include sales, traffic, conversion, ads-sp, ads-sb, ads-sd, ads-combined, refunds, profit-revenue, profit-cost, and profit-margin. Profit entries include isRequestable (only true entries can be requested), percentageMetric for the returned signed net-revenue share, and unavailableGroupings for SKU restrictions. Presentation-only profit entries have no callable domain. `domains` lists which tool families accept the code: product, marketplace, keyword, searchQuery, profit.
Returns aggregated account-level KPIs per marketplace over a date range, with an optional period-over-period comparison block. USE THIS FOR: "how is my business doing on Amazon DE last month", "sessions and conversion rate per marketplace this month", "compare revenue across all my marketplaces last quarter vs the quarter before", "total ad spend across all marketplaces last week". IMPORTANT: Pass fromComparison AND toComparison together for period-over-period questions, otherwise the response has no comparison block. Omit marketplaceCodes to aggregate across every enabled marketplace. For time-series trends use get_marketplace_trends; for per-product breakdowns use get_product_kpis. Date ranges are capped at 730 days. Returns: One row per marketplace plus a grand total row marked with `isTotalRow: true` (always appended when data is present, regardless of marketplaceCodes filter). `meta.totalEntries` counts only marketplace rows, not the total row, so `data.length` may equal `meta.totalEntries + 1`. When a comparison range is supplied, `comparison` follows `data` row for row, so `comparison[i]` is always the same marketplace as `data[i]`; each row also carries its own marketplaceCode if you prefer to join by key, and a marketplace with no data in the comparison window appears with null KPI values. Response uses the English KPI label (e.g. "Units Ordered") as the JSON key, not the code; pass codes from get_kpi_definitions in input fields. If `meta.warnings` is present, the workspace has a data problem that affects the KPIs you just asked for — a history import still running, a disconnected Amazon account, or an expired connection. Read the warning and qualify your answer instead of presenting these figures as final. Call get_workspace_info for the full readiness picture.
Returns time-series account-level KPIs per marketplace, aggregated by day, week, or month. USE THIS FOR: "chart my revenue trend across marketplaces", "show me daily ad spend this month", "how has total ACOS moved over the last quarter", "graph sessions per marketplace per week last quarter". IMPORTANT: Use granularity DAY for ranges under ~30 days, WEEK for quarters, MONTH for longer spans. For aggregated totals with comparison use get_marketplace_kpis instead. Omit marketplaceCodes to include every enabled marketplace. Date ranges are capped at 730 days. Returns: One row per marketplace plus a total row marked with `isTotalRow: true`, each carrying a timeSeries array aligned to meta.headerDates. `meta.totalEntries` counts only marketplace rows, so `data.length` may equal `meta.totalEntries + 1` when the total row is present. Response uses the English KPI label (e.g. "Units Ordered") as the JSON key inside each timeSeries bucket, not the code; pass codes from get_kpi_definitions in input fields. If `meta.warnings` is present, the workspace has a data problem that affects the KPIs you just asked for — a history import still running, a disconnected Amazon account, or an expired connection. Read the warning and qualify your answer instead of presenting these figures as final. Call get_workspace_info for the full readiness picture.
Returns product KPIs ranked by `sortBy` over a date range, with an optional period-over-period comparison block. Defaults to ranking the whole workspace catalog; narrow the ranking with `asinMarketplaces[]` (ASIN+marketplace tuples; each tuple selects one whole product family, even when it names a child ASIN) and/or `marketplaceCodes[]` (marketplace scope). Both filters can be combined: rows must match a tuple AND fall in one of the marketplace codes. USE THIS FOR: "top 10 products by revenue last month", "how is product B07XYZ1234 performing on amazon.de" (asinMarketplaces with one tuple, which returns that product family), "revenue and units for these specific ASINs", "best performers on amazon.de and amazon.com", "compare this month vs last month for these products". IMPORTANT: Pass fromComparison AND toComparison together for period-over-period questions, otherwise the response has no comparison block. For time-series charts use get_product_trends instead. A child ASIN resolves to its whole product family, so a tuple never returns the child alone. To see one child ASIN, set groupBy "child" and read the row for that ASIN. Date ranges are capped at 730 days. Returns: `data` — one row per product family (or per child ASIN with groupBy "child") with KPI values, ranked by sortBy. `totalRow` — the aggregate across every row matching the filter, not just the current page; use it for totals instead of summing rows, which is outright wrong for rate and ratio KPIs. When a comparison range is supplied: `comparison` covers exactly the products in `data`, in the same order, so `comparison[i]` is always the same product as `data[i]`; each row also carries its own asin and marketplaceCode if you prefer to join by key. A product with no data in the comparison window appears as a row with null KPI values. `comparisonTotalRow` is the comparison-window equivalent of `totalRow`. Meta includes timezone, currency, KPI set, sortBy, and resolved date ranges. Response uses the English KPI label (e.g. "Units Ordered") as the JSON key, not the code; pass codes from get_kpi_definitions in input fields. If `meta.warnings` is present, the workspace has a data problem that affects the KPIs you just asked for — a history import still running, a disconnected Amazon account, or an expired connection. Read the warning and qualify your answer instead of presenting these figures as final. Call get_workspace_info for the full readiness picture.
Returns time-series KPI rows for one or more products over a date range, aggregated by day, week, or month. USE THIS FOR: "chart my revenue trend", "show me sessions per day for my top five products this month", "how has product X performed over the last quarter", "graph units ordered per week for these three ASINs". IMPORTANT: Use granularity DAY for ranges under ~30 days, WEEK for quarters, MONTH for longer spans. Time-series tools return ONE metric per series — pass a single `kpi` code; for aggregated totals or multi-KPI questions with period-over-period comparison use get_product_kpis instead. Defaults to charting the whole workspace catalog (top rows by `sortBy` plus a total row, paginated by `page` and `pageSize`); narrow with `asinMarketplaces[]` (ASIN+marketplace tuples; each tuple selects one whole product family, even when it names a child ASIN) and/or `marketplaceCodes[]` (marketplace scope). When both filters are supplied they intersect (rows must match a tuple AND a marketplace code). Use `groupBy: "parentChild"` (default) for one rolled-up series per product family, or `groupBy: "child"` for one series per ASIN (parent + each child variation). To see one child ASIN, set groupBy "child" and read the series for that ASIN. Date ranges are capped at 730 days. Returns: One row per product, each carrying a timeSeries array aligned to meta.headerDates. When multiple products are returned, a total row marked with `isTotalRow: true` is appended. `meta.totalEntries` counts only product rows, so `data.length` may equal `meta.totalEntries + 1` when the total row is present. Response meta.kpi carries the English label (e.g. "Units Ordered"), not the code; pass the code from get_kpi_definitions in the `kpi` input field. If `meta.warnings` is present, the workspace has a data problem that affects the KPIs you just asked for — a history import still running, a disconnected Amazon account, or an expired connection. Read the warning and qualify your answer instead of presenting these figures as final. Call get_workspace_info for the full readiness picture.
Returns product catalog entries — ASINs, parent/child relationships, names, marketplace, and SKUs. Two modes: free-text discovery via `searchText` (matches name, ASIN, parentAsin, SKU), or batch metadata hydration via `asinMarketplaces[]` (look up the exact catalog rows for a known list of {asin, marketplaceCode} tuples). Pair `asin` with `marketplaceCode` to identify a single listing for downstream tools that take `asinMarketplaces[]`. USE THIS FOR: "what products do I have", "find product B07XYZ1234", "look up the product for SKU X", "hydrate names/images for these {asin, marketplaceCode} tuples I already have". IMPORTANT: The same ASIN on multiple marketplaces is multiple distinct products — pair `asin` with `marketplaceCode`. Supply `searchText` OR `asinMarketplaces[]`, not both; `asinMarketplaces[]` and `marketplaceCodes[]` are also mutually exclusive (tuples already include a marketplace), and `asinMarketplaces[]` matches listings 1:1 with no parent/child family expansion (tuples that do not map to a product are silently omitted). For listing issues use list_events with eventCategories ["listings-item-issues-change-event"]; for KPI numbers use get_product_kpis. Soft-deleted products are excluded by default; pass includeDeleted to include them. Returns: Product list with id, asin, parentAsin, isParent, name, marketplaceCode, mainImage, and SKUs. Filterable by searchText, asinMarketplaces, marketplaceCodes, and includeDeleted.
Returns the ASINs that rank for one tracked keyword on one marketplace, with their product KPI snapshot over a date range. USE THIS FOR: "for keyword X, which of my products rank and how do they perform", "how do my products that target keyword Y compare on revenue and units sold", "which ASINs win the keyword <phrase> on DE". IMPORTANT: The keyword is the free-text LABEL a shopper types (NOT an ObjectId), matched case-insensitively against keywords already tracked in this workspace — if the lookup returns "Keyword not found", the user must add it to keyword tracking first. This tool returns sales and traffic KPIs per ASIN, not rankings. For keyword rank history or rank KPIs use get_keyword_trends or get_keyword_kpis; their ranks are averages across the whole product family, so there is no per-ASIN rank. Date ranges are capped at 730 days. Returns: One row per ranking ASIN with name, brand, mainImage, marketplaceCode, and KPI values. Response uses the English KPI label (e.g. "Units Ordered") as the JSON key, not the code; pass codes from get_kpi_definitions in input fields. If `meta.warnings` is present, the workspace has a data problem that affects the KPIs you just asked for — a history import still running, a disconnected Amazon account, or an expired connection. Read the warning and qualify your answer instead of presenting these figures as final. Call get_workspace_info for the full readiness picture.
Brand-level SQPR is disabled because report data is unavailable. Brand SQPR KPI codes cannot be requested; ASIN SQPR and general search-query metrics remain available. Returns the ASINs that show up under one Amazon search query on one marketplace, with their SQPR KPI snapshot over a date range. USE THIS FOR: "for search query <phrase>, how do my products perform", "which of my ASINs win the search query <phrase> on DE", "which products convert best for the search term <phrase>". IMPORTANT: The searchQuery is the free-text phrase a shopper typed (NOT an ObjectId), matched exactly against the Amazon SQPR report. Granularity is required and restricted to "weekly" or "monthly" — Amazon does not publish SQPR data daily. The kpis array must use SQPR keys; standard product KPIs from the "sales" or "ads-*" categories are not valid here. For SQPR metrics at the query level across many queries, use get_sqpr_kpis instead. Date ranges are capped at 730 days. Returns: One row per ASIN with name, brand, mainImage, marketplaceCode, and KPI values. Response uses the English KPI label (e.g. "Search Query Impressions") as the JSON key, not the code; pass codes from get_kpi_definitions in input fields. `meta.effectiveDateRange` is the range actually queried: Amazon publishes SQPR per whole week or month, so the requested range is widened to whole report periods. Quote that range, not the one you asked for.
Breaks down the expandable profit categories into their detailed components: gross revenue, refunds, referral fees, FBA fees, other Amazon fees, shipping, gift wrap, advertising, and other costs. It uses the same calculations as the profit dashboard. USE THIS FOR: "show my profit cost breakdown", "which Amazon fees make up my costs", "what is inside my refund or advertising costs". To answer "why is my margin down" or to explain the difference between revenue and profit, call this tool together with get_profit_summary. IMPORTANT: from and to are timestamps in the user's local time with the correct UTC offset, not dates, covering whole local days (00:00:00 to 23:59:59); see the field descriptions. This tool returns only the expandable categories listed above. COGS and the revenue and profit subtotals never appear in it. VAT and promotions appear only when part of them is estimated for pending orders Amazon has not priced yet; they are then split into a from-Amazon child and an estimated child. Use get_profit_summary for COGS, VAT and the revenue-to-profit totals, and combine both tools to explain a margin. Only an unfiltered, ungrouped profit summary carries the full account cost scope. SKU rows cannot state ad spend: profitAdSpend, profitDb3 and profitNetProfit are null per SKU; profitDb2 is the available bottom line at that grain. Product filters exclude costs that cannot be attributed to the selected products, including account-level and marketplace-level costs. An explicit marketplace filter also excludes overall user-entered costs. Grouped rows may not sum to the summary: preserve dataUnassigned and use the grand total instead of summing rows. Check meta.isComplete, meta.unattributableKpis and meta.hasProductFilter before stating a total. isComplete describes cost attribution only, not report coverage or whether COGS and other cost settings are maintained; these are not verified. Profit KPIs include overlapping subtotals and fee components, so do not add all KPI columns together; use profitNetProfit for net profit and get_profit_breakdown for the cost composition. Refunds use period basis. No comparison period is provided. Parent totals and their children use the same dashboard calculation. isCredit describes the direction of a child; credit children add to revenue and offset costs. Child percentages are signed shares of net revenue, or null when net revenue is zero or negative (for example a product filter whose refunds exceed its sales). The remainder is already represented by a remainder child where applicable; do not add it again. Raw Amazon fee types and user categories have descriptive labels but no catalog code. Date ranges are capped at 730 days. Returns: A data object keyed by English parent labels. Each parent has children, remainder, and total. Children have label, value, isCredit, percentage, and code only when the child is a catalog KPI. Meta includes currency, dateRange (the from and to timestamps as sent), period refund basis, cost attribution completeness, and available warnings.
Returns profit KPIs ranked by SKU, child ASIN, parent ASIN or marketplace using the dashboard profit overview. USE THIS FOR: "which SKU loses money", "profit by product", "compare marketplace profitability". Use get_profit_summary for account totals and get_profit_breakdown for cost categories. IMPORTANT: from and to are timestamps in the user's local time with the correct UTC offset, not dates, covering whole local days (00:00:00 to 23:59:59); see the field descriptions. Only an unfiltered, ungrouped profit summary carries the full account cost scope. SKU rows cannot state ad spend: profitAdSpend, profitDb3 and profitNetProfit are null per SKU; profitDb2 is the available bottom line at that grain. Product filters exclude costs that cannot be attributed to the selected products, including account-level and marketplace-level costs. An explicit marketplace filter also excludes overall user-entered costs. Grouped rows may not sum to the summary: preserve dataUnassigned and use the grand total instead of summing rows. Check meta.isComplete, meta.unattributableKpis and meta.hasProductFilter before stating a total. isComplete describes cost attribution only, not report coverage or whether COGS and other cost settings are maintained; these are not verified. Profit KPIs include overlapping subtotals and fee components, so do not add all KPI columns together; use profitNetProfit for net profit and get_profit_breakdown for the cost composition. Refunds use period basis. No comparison period is provided. Returns: data contains grouped identities (ASIN, SKU, ISO marketplace code where applicable), kpis and signed percentages of net revenue, both keyed by English KPI labels; totalRow contains the grand total across all matching pages. dataUnassigned retains costs no detail row can claim. Do not add the residual to totalRow: it is already included. Meta includes pagination, sortBy as an input KPI code, groupBy, attribution completeness and warnings. Product-family filters include children; implicit denominator columns are omitted unless requested.
Returns account profit, revenue, Amazon fees, COGS, advertising and other costs for an inclusive time range given as ISO 8601 timestamps with a UTC offset (the user's local time), using the same calculations as the profit dashboard. USE THIS FOR: "what did I earn last month", "my total costs", "net profit and margin". For the cost waterfall use get_profit_breakdown; for product ranking use get_profit_by_product. IMPORTANT: from and to are timestamps in the user's local time with the correct UTC offset, not dates, covering whole local days (00:00:00 to 23:59:59); see the field descriptions. Only an unfiltered, ungrouped profit summary carries the full account cost scope. SKU rows cannot state ad spend: profitAdSpend, profitDb3 and profitNetProfit are null per SKU; profitDb2 is the available bottom line at that grain. Product filters exclude costs that cannot be attributed to the selected products, including account-level and marketplace-level costs. An explicit marketplace filter also excludes overall user-entered costs. Grouped rows may not sum to the summary: preserve dataUnassigned and use the grand total instead of summing rows. Check meta.isComplete, meta.unattributableKpis and meta.hasProductFilter before stating a total. isComplete describes cost attribution only, not report coverage or whether COGS and other cost settings are maintained; these are not verified. Profit KPIs include overlapping subtotals and fee components, so do not add all KPI columns together; use profitNetProfit for net profit and get_profit_breakdown for the cost composition. Refunds use period basis. No comparison period is provided. Returns: data is a flat object keyed by English KPI labels; percentages gives the corresponding signed share of net revenue for currency KPIs. Null values remain unknown. Meta includes currencyCode, dateRange (the from and to timestamps as sent), refundBasis, attribution completeness and import warnings. Percentage twins are returned here, not accepted as KPI input columns.
Brand-level SQPR is disabled because report data is unavailable. Brand SQPR KPI codes cannot be requested; ASIN SQPR and general search-query metrics remain available. Returns Amazon Search Query Performance Report metrics at the query level — search volume, impressions, clicks, cart adds, orders, conversion rate, and your share within each metric. USE THIS FOR: "what search queries drive my impressions", "top converting search terms last month", "queries where I lose share to competitors", "search queries that brought traffic to ASIN B07XYZ1234". IMPORTANT: SQPR is only published weekly and monthly — daily granularity is not available. SQPR data only exists at the child-ASIN level (parent ASINs have no SQPR rows). Filter to specific products with `asinMarketplaces[]` tuples; without filters returns the workspace-wide top queries. Use `groupBy: "searchTerm"` (default) for one row per (searchQuery, marketplace) aggregating across all matching child ASINs, or `groupBy: "asin"` to break out one row per (searchQuery, marketplace, asin) — the per-asin view. ASIN-share and ASIN-level absolute KPIs only make sense when a product filter is supplied — without filters, prefer workspace totals and computed rates. For time series use get_sqpr_trends instead. Date ranges are capped at 730 days. Returns: With groupBy="searchTerm" — one row per (searchQuery, marketplaceCode). With groupBy="asin" — one row per (searchQuery, marketplaceCode, asin). Response uses the English KPI label (e.g. "Search Query Impressions") as the JSON key, not the code; pass codes from get_kpi_definitions in input fields. `meta.effectiveDateRange` is the range actually queried: Amazon publishes SQPR per whole week or month, so the requested range is widened to whole report periods. Quote that range, not the one you asked for.
Brand-level SQPR is disabled because report data is unavailable. Brand SQPR KPI codes cannot be requested; ASIN SQPR and general search-query metrics remain available. Returns time-series Amazon Search Query Performance Report metrics, aggregated by week or month. USE THIS FOR: "chart search query volume for term X", "how has my click share trended this quarter", "weekly conversion rate by search query", "graph monthly impressions for queries that touch ASIN B07XYZ1234". IMPORTANT: SQPR granularity is restricted to WEEK or MONTH — daily is not available. SQPR data only exists at the child-ASIN level (parent ASINs have no SQPR rows). Pick one KPI per call; for multi-KPI or aggregated totals use get_sqpr_kpis instead. Filter to products with `asinMarketplaces[]` tuples. Use `groupBy: "searchTerm"` (default) for one series per searchQuery aggregating across all matching child ASINs, or `groupBy: "asin"` to break out one series per (searchQuery, asin) — the per-asin view. ASIN-share and ASIN-level absolute KPIs only make sense when a product filter is supplied. Date ranges are capped at 730 days. Returns: With groupBy="searchTerm" — one row per searchQuery. With groupBy="asin" — one row per (searchQuery, asin). Each row carries a timeSeries array aligned to meta.headerDates. Response meta.kpi carries the English label (e.g. "Search Query Impressions"), not the code; pass the code from get_kpi_definitions in the `kpi` input field. `meta.effectiveDateRange` is the range actually queried: Amazon publishes SQPR per whole week or month, so the requested range is widened to whole report periods. Quote that range, not the one you asked for.
Returns the workspace grounding state: name, currency, current date in UTC, the enabled Amazon marketplaces (ISO code + region), and known connection or import problems. Includes earliest and latest source-report dates observed in ClickHouse per marketplace. Continuous coverage and final attribution are not verified. USE THIS FOR: "what marketplaces does this workspace cover", "what currency are you reporting in", resolving relative date phrases like "last week" or "this month" to ISO dates before calling any date-taking tool, and checking data readiness before you state any figure as fact. IMPORTANT: Call this BEFORE reporting numbers from this workspace, not only for date grounding. A workspace can be mid-import, have an expired Amazon connection, or have no Ads account attached — in all three cases the other tools still return numbers, and those numbers may be incomplete while still looking plausible. The Warnings section tells you which KPI domains (sales, traffic, ads, profit) are affected. The profit domain is affected while the profit history import is still running, and also when the Selling Partner API or Ads API is not connected, the Selling Partner API connection has expired or onboarding is unfinished. If a domain you are about to report on is listed, say so in your answer instead of presenting the figure as final. None of the other tools parse "last 7 days" / "this quarter"; resolve such phrases to explicit dates first. Every tool except the three profit tools takes YYYY-MM-DD dates that are UTC calendar days (Sellerfox supports only UTC days for this data), so do not shift dates to the user's local time; currentDate is today's date in UTC. get_profit_summary, get_profit_breakdown and get_profit_by_product instead take ISO 8601 timestamps in the user's local time with the correct UTC offset, covering whole local days (00:00:00 to 23:59:59). Other tools accept ISO marketplace codes ("de", "us", "gb"), not Amazon internal marketplace IDs. For KPI code meanings call get_kpi_definitions, not this tool. Returns: A markdown text block with workspace name, currency, currentDate (the current date in UTC), enabled marketplaces (each with code and region), a Data Readiness section (Selling Partner API status, Ads API status, and the progress of any history import still running), and a Warnings section listing each data-quality problem with its severity and the KPI domains it affects. structuredContent carries the same facts as workspace, marketplaces, readiness and warnings, plus dataAvailability: the source dates with coverageStatus unverified and nullable firstReportDate/latestReportDate per source and marketplace.
Returns the workspaces this MCP connection can query. The LLM should call this first when the user mentions multiple brands/accounts to discover which workspaceId to pass to other tools. USE THIS FOR: "which brands am I connected to", "list my workspaces", "what accounts can you see", routing a multi-workspace question to the right workspaceId before calling any workspace-scoped tool. IMPORTANT: The returned workspaceId is the value to pass as the workspaceId argument to every other tool. For grounding metadata on a specific workspace (currency, current UTC date, enabled marketplaces) call get_workspace_info with that workspaceId. This tool takes no parameters — eligibility is derived from the current MCP connection. Returns: { data: [{ workspaceId, name, currency }] } — an object with a `data` array of rows for the workspaces this connection is authorized to query.
Returns business events that explain performance changes — buy-box losses, price changes, listing status changes, catalog edits, listing issues, and custom events. USE THIS FOR: "why did my sales drop", "what happened to ASIN B07XYZ1234 around date Y", "show me listing issues this week", "any buy-box losses last month on DE". IMPORTANT: Events overlap a date range if their [startAt, endAt] interval intersects it — a multi-day event that started before the range still appears. Filter by eventCategories[] for specific types (e.g. ["listings-item-issues-change-event"] for listing issues, ["buy-box-lost-event"] for buy-box losses); passing an invalid value returns the full valid list in the error message. Filter to products with asin + marketplaceCode. Date ranges are capped at 730 days. Returns: Event list with date range, affected products, event type, and event-specific log payload. Categorised via the discriminator pattern.
Lists manual and Amazon-derived income and expense entries with their category and allocation scope. USE THIS FOR: "what did I book in July", "show me recurring expenses", "which storage fees did Amazon charge", "find my hosting expenses". IMPORTANT: Only an unfiltered, ungrouped profit summary carries the full account cost scope. SKU rows cannot state ad spend: profitAdSpend, profitDb3 and profitNetProfit are null per SKU; profitDb2 is the available bottom line at that grain. Product filters exclude costs that cannot be attributed to the selected products, including account-level and marketplace-level costs. An explicit marketplace filter also excludes overall user-entered costs. Grouped rows may not sum to the summary: preserve dataUnassigned and use the grand total instead of summing rows. Check meta.isComplete, meta.unattributableKpis and meta.hasProductFilter before stating a total. isComplete describes cost attribution only, not report coverage or whether COGS and other cost settings are maintained; these are not verified. Profit KPIs include overlapping subtotals and fee components, so do not add all KPI columns together; use profitNetProfit for net profit and get_profit_breakdown for the cost composition. Refunds use period basis. No comparison period is provided. This ledger is only one part of profit tracking, so meta.isComplete is always false; use get_profit_summary for totals. Amounts retain each entry's original currency and are positive magnitudes; type identifies income or expense. Omit source to include both manual and Amazon-derived entries. Legacy entries without a source count as user entries. Supply both from and to or omit both. Ledger dates use UTC calendar days, matching the dashboard ledger, and are capped at 730 days. Search text is literal, case-insensitive title matching. Category, transaction, template, product, and entry IDs are never exposed. Returns: A paginated list with date, title, type, net amount, tax, original currency, available category name, source, appliesTo scope, marketplaceCodes when scoped, and isRecurring derived from a recurring template. Missing or deleted categories have no categoryName. Sorting by amount uses signed gross (net + tax), with expenses negative. Meta includes paging, timezone (UTC), optional dateRange, period refund basis, and completeness warnings.
sellerfox ChatGPT Plugin FAQ
How the directory, categories and Discoverability Score work.
Read the methodologyHow do I improve sellerfox's ChatGPT Plugin discoverability?
The levers are the listing surface agents actually read: names, descriptions, keywords, tool metadata, and registry health. Which lever matters depends on where discovery breaks, which is what continuous measurement shows.
Where is this profile measured?
This profile uses the geography attached to the latest public registry snapshot: US. Locale tags are intentionally omitted.