- Brand
- Agent Central
- Category
- AI
- Primary Subcategory
- Marketplace Seller, Dropship & Wholesale Tools
Integration details
Description
Agent Central connects the AI you already use to your Amazon Seller Central account. In ChatGPT, view business reporting for authorized Amazon sellers: campaign performance, inventory, orders, catalog details, fees and financial events, and fulfillment facts. The connection is view-only and limited to the selected authorized account and scope. Results reflect Amazon and Agent Central records; freshness and coverage depend on the connected sources.
- Integration type
- Plugin
- Verification status
- Not applicable
- Platform
- ChatGPT
- Primary Subcategory
- Marketplace Seller, Dropship & Wholesale Tools
- Secondary Subcategories
- None listed
- Brand
- Agent Central
- Access
- Account required
- First tracked
- 2026-11-22
- Tool count
- 66
- Market
- US
The Primary Subcategory used for this profile’s headline score.
Other Subcategories where the Integration is listed.
Not discoverable in ChatGPT yet
ChatGPT now suggests Plugins on its own when they match a user's request, but not every Plugin is being surfaced yet.We haven’t seen Agent Central suggested yet. Sign up and we’ll track it, and your Discoverability Score will appear here once ChatGPT starts surfacing it.
What discovery looks like

Get alerts for Agent Central
Get updates when Agent Central’s Discoverability Score or category rank changes.
Competing in ChatGPT Marketplace Seller, Dropship & Wholesale Tools
View Category66 tools agents can invoke
Use for raw AWD (Amazon Warehousing & Distribution) inventory snapshot: on-hand and inbound quantities per SKU. AWD inventory snapshots do not expose FNSKU in this dataset. Do not use for FBA fulfillable stock or ASIN-level identity; use get_fba_inventory or get_listing_registry with view='listings'. Included in get_days_of_cover.
get_awd_inventory
Get per-SKU line items from AWD shipments — SKU, quantity per shipment. Use to prorate AWD/AGL freight charges to specific ASINs. Defaults to latest snapshot.
get_awd_shipment_items
Get AWD shipment-level data — shipment IDs, statuses, and timestamps. For per-SKU quantities within a shipment, use get_awd_shipment_items.
get_awd_shipments
Use to drill Sponsored Products performance from campaign to ad group: spend, sales, ACOS, ROAS, CTR, CPC, CVR. Grain: campaign + ad group; recent ~3 days may be incomplete. Do not use for bidded keyword/target or customer search-term analysis; use get_keyword_performance or get_search_terms. Each row's refresh_window_open is true when at least one included reporting date is on or after settling_from_date and on or before reporting-local today. settling_from_date is the start of that profile/table's latest completed scheduled re-pull window (normally 16 calendar days ending the day before refresh). It means we still re-pull the day, not that its value will change; false is not a guarantee of finality. Missing completed-window evidence returns null.
get_adgroup_performance
Returns Amazon's console change-history events for Sponsored Products and Sponsored Brands (the Seller Central History tab, last 90 days). IMPORTANT: changes made through the API, including every change agentcentral makes, do NOT appear here, so this is typically empty for agent-managed accounts. To see changes agentcentral made use get_action_history; to detect any Sponsored Products campaign-setting change regardless of source use get_campaign_snapshots with view=changes. Amazon does not include who made the change and does not return Sponsored Display changes.
get_ads_change_history
Use when the user asks to compare Amazon sponsored advertising performance between two periods. Returns changes across campaign, ad, search-term, targeting, or advertised-product dimensions supported by the selected ad product. Defaults to Sponsored Products; also supports Sponsored Brands and Sponsored Display. Uses stored reports, not live entity state or write verification. Comparisons show observed changes, not proof of causation.
diagnose_performance
Use for Amazon demand-side platform (DSP) Reporting v3 performance at order, line item, and creative grain: spend, sales, ROAS, viewability, and video completion. Do not use for Sponsored Products/Brands/Display, search terms, or seller account sales; use the matching ads or sales tool.
get_dsp_performance
Use when the user requests an overview of their connected Amazon seller account. Returns stored sales, advertising, inventory, top-selling products, period comparisons, data freshness, and coverage summaries where available. Requires Ads or Inventory access. Read-only; does not change Amazon records or prepare write actions.
get_account_data_tour
Use for Amazon Attribution tagged off-Amazon traffic, with publisher, campaign/ad group/creative, product, purchases, sales, brand halo, and new-to-brand metrics. Product fields product_asin, product_name, brand_name, and marketplace are populated for report_type=PRODUCTS; PERFORMANCE rows can leave them null. Do not use for on-Amazon ads, search terms, or total seller sales; use campaign/search/sales tools.
get_attribution_performance
Get Amazon Ads Brand Metrics health by brand and category: awareness, consideration, sales, engaged-shopper rate, conversion rate, and new-to-brand rate versus category peers.
get_brand_health
Use when the user requests Amazon Brand Analytics Search Terms reports: top_terms returns search-term ranks, term_products returns clicked products for one exact term, and asin_terms returns terms for one exact clicked ASIN. Reads stored reports for a required date window of at most 42 days. Lookup views require the exact term or ASIN; wildcard term searches are not supported.
get_brand_search_terms
Use for SP, SB, or SD budget usage and pacing snapshots: Amazon's near-real-time usage percentage plus derived estimated used and remaining budget by campaign. The omitted ad_product default remains SP. Do not treat estimated used budget as finalized advertising cost; use retained campaign performance for finalized spend.
get_budget_pacing
Use for Sponsored Products campaign-level ad performance over a date range: spend, sales, ACOS, ROAS, CTR, CPC, CVR. Grain: one row per campaign; recent ~3 days may be incomplete. Do not use for ad group, keyword, search-term, ASIN, or total organic+paid sales questions. Each row's refresh_window_open is true when at least one included reporting date is on or after settling_from_date and on or before reporting-local today. settling_from_date is the start of that profile/table's latest completed scheduled re-pull window (normally 16 calendar days ending the day before refresh). It means we still re-pull the day, not that its value will change; false is not a guarantee of finality. Missing completed-window evidence returns null.
get_campaign_performance
Use for Sponsored Products campaign settings snapshots from the campaign_snapshots sync. view='history' returns daily state, budget, bidding, portfolio, and placement bid adjustment rows. view='changes' returns observed setting differences between daily snapshots, including changes made in the console or through the API. It requires at least two days of snapshots and cannot show intermediate changes between captures. This is not a live write readback and does not return performance metrics; use get_campaign_performance for metrics.
get_campaign_snapshots
Use when the user asks about the current connection, marketplace, account access, available date ranges, or enabled tools. Returns the key's display name and safe prefix, marketplace scope, connection status, source coverage dates, historical access limit, and effective tool access. api_key.permissions lists special capabilities; tools and tool_scope_by_domain describe effective read/write access. view=data_map returns source and tool mappings, not business performance data.
get_account_info
Use for per-ASIN/per-FBA-SKU replenishment coverage: days of stock remaining computed from FBA fulfillable + FBA inbound divided by recent FBA sales velocity. Uses completed Sales & Traffic Business Reports when every due date in the lookback is loaded (the newest dates inside Amazon's finalization lag are reported as not yet due and excluded from the velocity average); when initial Business Report backfill is incomplete and the lookback is 7/30/60/90 days, uses Amazon FBA Inventory Health shipped-unit velocity as a source-attributed fallback and surfaces the sales-history data gap. AWD on-hand is reported as its own field but excluded from the cover numerator (AWD is upstream of FBA, not directly fulfillable). Sorted by lowest cover first. Do not use for raw stock counts, aged inventory fees, or broad triage; use get_fba_inventory/get_inventory_health/get_inventory_risk_triage.
get_days_of_cover
Use for stored Fulfillment by Amazon (FBA) inventory quantities per ASIN/SKU: fulfillable, reserved (including customer-order, FC-transfer, and FC-processing breakdowns), inbound, future supply, researching, and unfulfillable. Future supply fields are Amazon source-provided In-Stock Head Start quantities and remain null when Amazon does not return them. Defaults to latest snapshot. Returns stock quantities, not sales velocity or days-of-cover calculations.
get_fba_inventory
Returns stored monthly FBA storage estimates, Amazon Seller Economics product allocations with source SKUs, or long-term storage charges, as source rows or summed per product or per month. Source labels distinguish allocations from posted settlement charges. Empty months report Amazon's request outcome. Reads never create reports. FBA only, not AWD.
get_fba_storage_fees
fee_breakdown supports exact event_type (Shipment, Refund, Adjustment, ServiceFee) and component_type filters in SQL before aggregation and pagination. For storage postings use grain='raw_component', event_type='ServiceFee', and component_type='FBAStorageBilling' or 'FBALongTermStorageBilling'; omit asin/order_id to retain account-level rows. ASIN/order-item pivots include only matching components and still exclude rows without an ASIN/order respectively. Filters are not valid for profitability. Return Finances v0 financial-event rows and deterministic ASIN economics. view='fee_breakdown' returns raw order components when order_id is supplied; otherwise grain='asin' returns the existing ASIN pivot, grain='order_item' returns bulk order/SKU/ASIN fee pivots, and grain='raw_component' returns component rows across the date range. view='profitability' returns ASIN-level shipment_principal, source-signed refund_principal, source-signed refund_net, revenue, fees, tax, tax_withheld, promotions, net, units, refund-event coverage, and data-gap fields. revenue remains shipment_principal + refund_principal for backward compatibility. Promotions and marketplace-facilitator tax Amazon withheld are separate signed components (component_type 'Promotion' and 'ItemWithheldTax:<charge>'; a repeat on one item is numbered, e.g. 'Promotion#2'): they are included in net and net_profit and never counted as fees. Its money fields are selected by financial-event posted_date, not order date; order_item_revenue is an independently filtered purchase-date reference, so posted-date economics must not be subtracted from order-date sales as though both cutoffs represented the same orders. Fees on off-Amazon fulfillment (MCF) orders, which carry no Principal, are excluded from total_fees and net_profit and returned in off_amazon_fulfillment_fees (net_including_off_amazon_fulfillment reconciles); prior_period_order_share reports how much posted shipment principal came from orders placed before start_date. Profitability includes Shipment/Refund events only and excludes all ad spend, FBA storage fees, inbound freight, COGS, and account-level Adjustment/ServiceFee events; responses expose date_basis, scope, excluded_costs, and reconciliation_note. Account-level Adjustment/ServiceFee rows remain available with view='fee_breakdown' and grain='raw_component', where they are surfaced for any marketplace key. Rows are attributed to the marketplace Amazon reports for each event, and a marketplace's rows can include components in another currency (for example cross-border fees), so money is scoped to the marketplace currency by default; set include_all_currencies=true to include all component currencies (amounts are then mixed across currencies and should not be summed). Rows carry data_gap='no_refund_events_in_period' when refund adjustments are absent. For pre-settlement Transaction View rows use get_payment_transactions; for finalized settlements use get_settlement_economics.
get_financial_events
Get FBA inbound shipment data — shipment IDs, statuses, destination fulfillment centers, and current-state refresh dates. snapshot_date is when a shipment's data was last refreshed. Amazon's legacy shipment-list response does not provide a creation timestamp; created_at is a best-effort inference from a date embedded in shipment_name, and created_at_source identifies that inference. Ambiguous unverified date formats and future inferred dates return null instead of a silently incorrect date. Amazon does not provide a per-shipment received or closed timestamp here; for when units were received at the fulfillment center use get_inventory_movements with event_type=Receipts (reference_id is the inbound shipment ID), or get_shipment_receiving_status for shipped-vs-received quantities per SKU. Inbound shipments are account/region-level data (no native marketplace): rows surface tenant-wide for any marketplace-scoped key and are anchored to the tenant's primary marketplace.
get_inbound_shipments
Per-fulfillment-center on-hand inventory. Returns the absolute ending warehouse balance per ASIN, SKU, fulfillment center (FC code like ONT8), and disposition, plus the day's signed flow components (receipts, customer shipments, customer returns, warehouse transfers, in-transit). Use when the user asks about per-warehouse or geographic distribution of inventory. Account-level inventory snapshots do not provide this warehouse breakdown. Source: Amazon Inventory Ledger Summary View (GET_LEDGER_SUMMARY_VIEW_DATA, aggregateByLocation:FC). Defaults to the most recent snapshot date. Ledger summary data is account/region-level (no native marketplace): rows surface tenant-wide for any key and are anchored to the tenant's primary marketplace.
get_inventory_by_fulfillment_center
Returns FBA inventory health snapshot fields: age buckets, capped weeks of cover, storage fee risk, sell-through, and Amazon source-provided order and ship-in quantities by SKU/ASIN. Reimbursement placeholder SKUs are filtered by default; set include_synthetic_skus=true to include rows flagged with is_synthetic. For raw FBA stock counts use get_fba_inventory; for cross-signal risk rankings use get_inventory_risk_triage.
get_inventory_health
Get inventory ledger movements — shipments, receipts, adjustments, customer orders, and returns by date/ASIN/SKU. reference_id is the source document for each movement: for Receipts events it is the inbound FBA shipment ID, so receipts can be joined to get_inbound_shipments (shipment_id) to estimate created-to-received time when created_at_source confirms a shipment-name inference. date_and_time is the event timestamp (UTC). Ledger movements are account/region-level data (no native marketplace): rows surface tenant-wide for any marketplace-scoped key and are anchored to the tenant's primary marketplace.
get_inventory_movements
Returns ranked inventory risk signals across FBA/AWD stock, sales velocity, inventory health, and suppressed-listing data, including days-of-cover and risk scores derived from the same coverage definition as get_days_of_cover. Grain: ASIN/SKU risk rows. Do not use for raw snapshots or movement audit; use inventory snapshot or ledger tools.
get_inventory_risk_triage
Returns bidded Sponsored Products keyword/target performance. Grain: keyword/target + campaign/ad group with spend, sales, ACOS, ROAS, and impression share; recent ~3 days may be incomplete. Does not include actual customer search queries or market-wide query volume; use get_search_terms or get_search_query_performance. Each row's refresh_window_open is true when at least one included reporting date is on or after settling_from_date and on or before reporting-local today. settling_from_date is the start of that profile/table's latest completed scheduled re-pull window (normally 16 calendar days ending the day before refresh). It means we still re-pull the day, not that its value will change; false is not a guarantee of finality. Missing completed-window evidence returns null.
get_keyword_performance
Return keyword rank snapshot data from the keyword rank table. view='positions' returns rank positions across a date range (defaults to the trailing 7 days), view='changes' compares two rank snapshots (defaults to the two most recent), and view='with_volume' joins a rank snapshot (defaults to the latest) to the SQP week containing that date for search volume and share metrics.
get_keyword_ranks
Return factual listing-content changes from agentcentral-retained snapshots. Diffs title, brand, description, product type, status/classification, browse classifications, item/package dimensions, main image, bullets, and image set from append-only Catalog Items history, plus backend generic_keyword from weekly Listings Items snapshots. Returns one row per changed field with source snapshot dates. Amazon exposes no native listing edit-history API; first-snapshot accounts report insufficient retained history instead of claiming no change. Video and A+ module content are not diffed.
get_listing_content_changes
Return the seller's own listing price economics per SKU/child ASIN from a Listings Items materialization, with completed per-SKU post-write refresh overlays when available. Compatibility scalars describe the ALL-audience price; structured purchasable_offer preserves all consumer and B2B schedules and quantity tiers, and listing_issues preserves Amazon issue code, severity, and enforcements. This does not make broad live SP-API calls; it is distinct from competitive/Buy Box pricing.
get_listing_offers
Return weekly Listings Items API quality snapshots. view='summary' returns listing completeness, issue counts, and quality status; view='issues' returns Amazon listing issue codes, affected attributes, and raw issue details. Reimbursement placeholder SKUs are filtered by default. Snapshots refresh weekly. Weekly snapshots do not establish current live buyability.
get_listing_quality
Use when the user asks about listing quality gaps across their catalog. Returns stored Amazon errors and warnings, missing core content, offer or fulfillment gaps, and content-depth signals by listing, sorted by deterministic cleanup score. Uses snapshots, not live listing validation, and does not edit listings.
get_catalog_cleanup
Return latest active listing and catalog data. view='listings' returns ASIN/SKU rows with available title, brand, product type, bullet points, description, backend generic keywords, Amazon listing creation date, price, quantity, and fulfillment fields; view='search' discovers or sorts the same listing rows by product/brand/title/SKU/ASIN text or creation date; view='brand_asins' resolves the active ASIN set for a catalog brand. Set include_content=false for short listings/search rows without listing copy. Content prefers Catalog Items data and falls back to Listings Items quality or seller active-listings data. Reimbursement placeholder SKUs are filtered by default.
get_listing_registry
Return weekly Amazon Brand Analytics Market Basket pair facts at primary-ASIN × paired-ASIN grain: Amazon purchased-with rank, combination percentage, report dates, and marketplace. Defaults to the latest complete Sunday-Saturday week. The tool does not label pairs as opportunities or recommend cross-sells.
get_market_basket
Use after you know an Amazon order_id to inspect one order's header and line items. Grain: single order with item-level ASIN/SKU/title/quantity/price and pallet_delivery_requirement. pallet_delivery_requirement is Amazon's buyer-selected per-item delivery constraint: mandatory means Amazon returned MANDATORY; not_required means FULFILLMENT shipping data was requested and did not require a pallet; unknown means legacy, report-backed, or unavailable/malformed data and is never ordinary-parcel confirmation. Amazon's Orders API omits monetary fields while payment is unauthorized; Pending FBA rows can be enriched from Amazon's order-tracking report and disclose item_price_source='amazon_order_tracking_report'. Unavailable prices are null, never a synthetic zero. When every line is priced, item_subtotal_amount exposes their exact sum; order_total_amount remains null until Amazon supplies the full order total, and Pending values remain non-final. purchase_date is the seller reporting date; purchase_at and purchase_date_utc are included for UTC audit. Do not use for order discovery or sales totals; use get_orders or get_sales_summary first.
get_order_details
Use to inspect line items across many orders in one call, filtered by seller reporting date, order status, ASIN, or fulfillment channel. Grain: one row per order line item (order_id, order_item_id, ASIN, seller SKU, title, quantities, prices, pallet_delivery_requirement). pallet_delivery_requirement is Amazon's buyer-selected per-item delivery constraint: mandatory means Amazon returned MANDATORY; not_required means requested FULFILLMENT shipping data did not require a pallet; unknown means legacy, report-backed, or unavailable/malformed data and is never ordinary-parcel confirmation. Amazon's Orders API omits monetary fields while payment is unauthorized; Pending FBA rows can be enriched from Amazon's order-tracking report. Check item_price_availability and item_price_source; unavailable monetary values are null, never a synthetic zero. Omitting start_date/end_date returns only the trailing 90 days ending today — for any prior period (a past month or last year, e.g. year-over-year) you MUST pass explicit start_date and end_date. The asin filter matches line items containing that ASIN (gross, including cancelled/zero-quantity orders), not units sold. purchase_date is the seller reporting date; purchase_at and purchase_date_utc are included for UTC audit. For per-ASIN aggregates (units/revenue over time) use get_sales_velocity; for order headers only use get_orders; for account daily totals use get_sales_summary.
get_order_items
Use to inspect or filter channel-complete individual order headers from the Orders API by seller reporting date, status, ASIN, or fulfillment channel. Grain: one row per order. Includes explicit Non-Amazon/Multi-Channel Fulfillment (MCF) channels and returns sales_channel on each row. Omitting start_date/end_date returns only the trailing 90 days ending today — for any prior period (a past month or last year, e.g. year-over-year) you MUST pass explicit start_date and end_date. The asin filter matches order headers containing that ASIN (gross, including cancelled/zero-quantity orders), not units sold. purchase_date is the seller reporting date; purchase_at and purchase_date_utc are included for UTC audit. Pending orders can have order_total_amount=null because Amazon has not finalized the order total yet. Individual order headers are not Amazon marketplace sales totals; use get_sales_summary for daily totals or product-level sales tools for ASIN/SKU breakdowns.
get_orders
Returns Sponsored Products performance by ad placement (Top of Search, Detail Page, Other, Off Amazon) within campaigns; recent ~3 days may be incomplete. Use to compare placement-level spend, sales, ACOS, ROAS, CTR, and CVR. Does not include campaign totals, keyword performance, or organic traffic; use get_campaign_performance/get_keyword_performance/get_sales_and_traffic.
get_placement_breakdown
Returns Sponsored Products performance rolled up by portfolio ID, joined from campaign metrics and latest portfolio assignment snapshot. Grain: one row per portfolio. For individual campaign/ad group/keyword metrics, use the matching drilldown tool.
get_portfolio_performance
Use for materialized product metadata for one ASIN: title, brand, product type, bullet points, description, backend generic keywords, images, item and package dimensions, status, and BSR. This is captured snapshot data, not a live Catalog Items response. Content prefers the latest captured Catalog Items row, then the latest marketplace-scoped Listings Items quality snapshot; seller listing-offer data can supply title and backend generic-keyword facts when those sources are absent. Source labels and per-source snapshot dates are returned so freshness is explicit. BSR comes in two grains: sales_rank/sales_rank_category is the most specific (sub)category rank Amazon reports, and main_category_rank/main_category is the top-level department rank (the headline '#N in <Department>' shown on the product page). Some attributes can be null when Amazon omits them from all captured sources. Do not use for seller SKU mapping, sales, inventory, or ad performance; use get_listing_registry with view='listings' or the relevant performance tool.
get_product_details
Return Amazon fee estimates from report-backed GET_FBA_ESTIMATED_FBA_FEES_TXT_DATA snapshots. view=current returns the latest snapshot, history returns bounded snapshot rows, and changes compares two compatible snapshots. Values are Amazon's estimates rather than actual-fee guarantees.
get_product_fee_estimates
Use for Sponsored Products advertised ASIN/SKU performance by campaign/ad group: SP spend, clicks, sales, ACOS, ROAS, same-SKU sales; recent ~3 days may be incomplete. This is not total ad spend for an ASIN or brand: it excludes Sponsored Brands, DSP, and any campaign spend not tied to the ASIN as the advertised product. Do not use for total organic+paid product sales or products bought after an ad click; use get_sales_and_traffic/get_tacos or get_purchased_products.
get_product_performance
Returns ranked ASIN profitability issue signals from Shipment/Refund financial-event net, FBA reimbursements, returns, Business Reports sales and units, and SP/SD product ad spend. Financial-event revenue, fees, tax, and marketplace net are selected by posted_date, not order date; Business Reports use their order-date basis, while ads, returns, and reimbursements use their own source dates. The same requested window can therefore contain different transaction cohorts near its cutoffs and must not be treated as one order-date cohort without accounting for posting lag. Account-level Adjustment/ServiceFee financial events are excluded because they cannot be deterministically attributed to an ASIN. Fees on off-Amazon fulfillment (MCF) orders, which carry no Principal, are excluded from fees, net, and contribution and returned in off_amazon_fulfillment_fees; prior_period_order_share reports how much posted principal came from orders placed before the period. Set include_sponsored_brands=true to return exact SB campaign spend once in summary.account_level_costs as an unallocated account-level cost. It is not allocated to, deducted from, or repeated across ASIN rows. Responses expose date_basis, scope, excluded_costs, and reconciliation_note. DSP ad spend, FBA storage fees, inbound freight, COGS, and account-level Adjustment/ServiceFee events remain excluded. Money is scoped to the marketplace currency by default; set include_all_currencies=true to include mixed currencies. A data gap reports when Refund financial events are absent in the period.
get_profitability_review
Returns Amazon promotion performance from synced coupon and deal data: view=coupon returns per-coupon clips, redemptions, budget used/remaining, discount, sales, dates, and covered ASINs from GET_COUPON_PERFORMANCE_REPORT; view=deal returns Best Deal and Lightning Deal glance views, units sold, revenue, Amazon-reported status, dates, and included products from GET_PROMOTION_PERFORMANCE_REPORT. Start/end dates, and the start_date/end_date filter, use the Selling Partner Promotions API schedule when a matching promotion exists, otherwise the Amazon report request window. Stats are cumulative values refreshed daily: the coupon view covers a rolling two-year start window (just inside Amazon's cap), the deal view a rolling 90-day window. Factual performance data only; no recommendations are made. Brand Tailored Promotions have no API and never appear here.
get_promotion_performance
Returns non-zero purchase attribution after ad clicks by default: advertised ASIN to purchased ASIN, campaign/ad group, sales, orders, and units; recent ~3 days may be incomplete. Set non_zero_only=false to include zero-attribution rows. Use for halo/cross-sell attribution patterns. Does not include spend, clicks, impressions, ACOS, or total product sales; use get_product_performance or get_sales_and_traffic/get_tacos.
get_purchased_products
Get FBA reimbursement records. Filter by date range, ASIN, or reason. Sorted by amount_total by default to surface highest-value reimbursements first. Reasons use Amazon's reimbursement report labels (Lost_Warehouse, Damaged_Warehouse, Lost_Inbound, CustomerReturn, CustomerServiceIssue, Reimbursement_Reversal, GeneralAdjustment); known Finances API descriptions on rows imported from that API are translated to the same labels; others are shown as received. Reimbursements are account/region-level FBA data (no native marketplace): rows surface tenant-wide for any marketplace-scoped key and are stored anchored to the tenant's primary marketplace.
get_reimbursements
Returns Brand Analytics repeat-purchase behavior by ASIN for synced week, month, or quarter report periods. Includes orders, unique customers, repeat-purchase revenue, and repeat-customer percentages. This is aggregate Brand Analytics data, not daily customer-level order identity.
get_repeat_purchase_behavior
Get FBA return records. Filter by date range, ASIN, SKU, return reason, or disposition status. Returns are account/region-level FBA data: rows surface tenant-wide for any marketplace-scoped key and are stored anchored to the tenant's primary marketplace (marketplace_id). The derived order_marketplace_id attributes each return to the marketplace of its matching order in order history, and is null when no matching order is found.
get_returns
Use for historical Business Report performance (~72h delay). For enabled accounts, view=account_daily returns source-provided account/day totals, including traffic-only days, without product attribution. The default products view returns: organic + PPC sales, sessions, page views, conversion, Buy Box, and the browser vs Amazon mobile-app traffic split (sessions, page views, and their percentages) by ASIN and source-provided seller SKU. In that view Amazon repeats a multi-SKU product's full sessions and page views on every SKU row and omits products with no orders that day, so never add traffic across rows; for per-product traffic use view=child_items where enabled (one row per child ASIN, no SKU), and for account traffic use view=account_daily. Seller SKU identity is case-sensitive; rows whose SKUs differ only by letter case are distinct source rows and should only be combined when intentionally rolling up to ASIN or account totals. Filter one variation family (all child ASINs of a parent) with parent_asin. Do not use for current-day/yesterday account totals or ad-attributed campaign metrics; use get_sales_summary/get_todays_sales or ads performance tools. Exact requested-date coverage is reported in sales_and_traffic_source_coverage; when gaps exist, do not infer continuous coverage from the table's outer data-window bounds. For account-level totals of historical weeks, get_sales_summary falls back to this same Business Report data when its newer sources (Sales API, Orders API) do not cover the range, so the two tools agree on whether a week has sales history.
get_sales_and_traffic
Use for latest Best Sellers Rank (BSR) across catalog ASINs, with rank, category, title, and brand. Each row carries two grains of BSR: sales_rank/sales_rank_category is the most specific (sub)category rank, and main_category_rank/main_category is the top-level department rank (the headline '#N in <Department>'). Grain: one row per ranked ASIN. Do not use for keyword rank tracking or actual sales velocity; use ranking tools or get_sales_velocity.
get_sales_ranks
Use for account-level daily Amazon marketplace sales totals, including yesterday or date ranges. Grain: one row per day with revenue, units, and orders. Final rows use Sales API daily metrics or deep-history Business Report daily rollups. Sales API snapshots take precedence; Business Report replaces same-date historical order-item fallback rows, while recent Orders rows retain precedence. Non-final recent/fallback rows use order-item rollups that exclude sales_channel values beginning Non-Amazon (MCF/off-Amazon fulfillment shipments). total_sales remains available priced line-item revenue only; Pending FBA prices are enriched from Amazon's order-tracking report when available, unpriced_unit_count remains separate, and estimated_total_sales adds unpriced_unit_count * average_unit_price when a priced day ASP exists. Canceled orders do not contribute priced or unpriced sales units. Historical rows can fall back to order-item rollups when the Sales API snapshot table has no row. Business Report rows use the same source as get_sales_and_traffic: source-provided account/day totals take precedence when stored, otherwise total_sales is ordered_product_sales summed across ASINs; these grains are never added together, order_item_count comes from the Business Report total_order_items, and order_count (distinct orders) is not available from that source and stays null. Read sales_channel_scope and unit_count_basis on each row; observed_order_count reconciles to get_orders only when the same date and Amazon-marketplace sales-channel scope is used. Check total_sales_basis, estimated_total_sales_basis, sales_completeness_state, is_final, revision_state, timezone_basis, and date_basis before comparing recent days to finalized baselines. Do not use for ASIN/SKU breakdowns, individual orders, or off-Amazon fulfillment volume; use get_todays_sales, get_sales_velocity/get_sales_and_traffic, get_orders, or fulfillment tools.
get_sales_summary
Use for historical Amazon-marketplace product sales velocity and inventory planning. Returns provisional order-item priced_revenue, priced_unit_count, unpriced_unit_count, currency and latest item snapshot alongside net_units_sold by ASIN/period from Orders API data plus Business Reports sales/traffic fields when available. Explicit Non-Amazon/MCF sales channels are excluded so order-derived units reconcile with get_sales_summary and Business Reports. Grain defaults to daily when asin is provided and per-ASIN range summary otherwise; set time_grain=day/week/month/summary for explicit periods. Do not use get_orders header counts as units sold; get_orders intentionally returns channel-complete raw order headers, including cancellations and MCF. Recent product revenue is limited to synced orders; missing prices are not estimated. Revenue excludes tax/shipping and remains separate from Business Report ordered_product_sales. Do not use for account-level totals; use get_todays_sales or get_sales_summary.
get_sales_velocity
Agent Central ChatGPT Plugin FAQ
How the directory, categories and Discoverability Score work.
Read the methodologyHow do I improve Agent Central'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.
What are Agent Central alternatives on ChatGPT?
As of 2026-11-22, Agent Central competes with Adzviser Amazon Seller Central, Adzviser eBay, AgentXray, Airsup Search, Amazon Product Research, AMZScout, AutoLister, Bonsai Solutions and 46 more in ChatGPT Marketplace Seller, Dropship & Wholesale Tools, ranked by public Discoverability Score.
Where is this profile measured?
This profile uses the geography attached to the latest public registry snapshot: US. Locale tags are intentionally omitted.