trackiq

Amazon MCP server — connect Claude, ChatGPT or Cursor to live Amazon Seller Central, Sponsored Ads, DSP, AMC, Search Query Performance, rank and inventory data through one Model Context Protocol connection.

Splits every Amazon Sponsored Products and Sponsored Brands search term into branded, competitor, product-targeting and generic buckets, then shows where the ad money really goes — how much is spent on shoppers who already typed the brand, what generic search returns once brand is taken out, whether competitor conquesting pays — and checks…

Tracks weekly organic rank movement across a brand's priority keyword set by sampling live search results, compares it to the previous run, and weights every gain and loss by what that keyword is worth in revenue so the drops that cost real money surface first and the rest stay quiet. Use when the user asks about rank movement,

Turns trailing sales and last year's seasonal shape into a week-by-week demand forecast per Amazon product, then converts it into order quantities and order-by dates that account for lead time, safety stock and the service level the brand wants — with the forecast error stated so nobody treats it as certain. Use when the user as

Reads Amazon Marketing Cloud's time-to-conversion data to show how long shoppers take between their first ad touch and purchase — the share that buys within the hour, the day and the week, and how that shifts month to month — then turns it into two decisions: which attribution window to report on, and how many days ahead of a sales event to…

TrackIQ: Amazon Share of Shelf — an Agent Skill measuring how much of Amazon page one a brand owns for its revenue-bearing keywords and who holds the rest. Weighted by revenue, not volume. Requires an Oxylabs scraper connection (1 credit per keyword per run).

TrackIQ: Amazon Listing Monitor — an Agent Skill that watches listings daily and alerts only when something actually moved: title and bullet rewrites, lost images, missing A+, category moves, buy box changes, suppressions. Silence is the normal output. Requires an Oxylabs scraper connection.

Agent Skills for Amazon sellers running the TrackIQ MCP. Install the whole catalog with one command in Claude Code, or upload any skill as a zip to Claude or ChatGPT.

Finds the Amazon search terms that already convert above break-even but have no keyword of their own, and turns them into an upload-ready Sponsored Products bulk file with exact-match keywords and suggested bids. Checks every candidate against the existing targets so nothing is proposed twice. Use when the user asks about harves

Breaks Amazon DSP down by audience, creative, inventory source, technology and geography to show which segments actually recruit customers and which burn impressions. Ranks on new-to-brand purchases, detail page views and cost per action rather than ROAS, because DSP revenue is not returned at these dimensions. Use when the user

Finds the Amazon campaigns still paying to advertise products that are about to run out of stock, and produces an upload-ready bulk file that throttles or pauses those ads with the real campaign, ad group and ad IDs attached — plus where the freed budget should go instead. Use when the user asks about advertising products that a

Builds a voice-of-customer picture for a brand by sampling reviews across its own catalogue and its competitors, clustering what buyers praise and complain about into themes, and turning those into bullet and A+ copy directions plus negative keywords. Uses the full star distribution for anything quantitative because only about e

Finds Amazon advertising spend that is not paying for itself — keywords that spent and sold nothing, keywords bidding above what they return, the same keyword duplicated across ad groups bidding against itself, and search terms worth negating — then produces an Amazon bulk-operations workbook with the campaign, ad group and keyw

Rewrites one Amazon listing — title, bullets, description and backend terms — against the terms it should rank for and the questions shoppers actually ask, producing a side-by-side of current and proposed copy with the reason for every change. Built for how Amazon matches intent and answers questions in 2026, not for keyword stu

Ranks an Amazon catalogue by how much each product's revenue moved between two periods and decomposes every mover into its cause — traffic, conversion rate, units per order, or average selling price — so the answer is "sessions fell 31%", not "sales are down". Use when the user asks why sales are down or up, what moved, which pr

Measures what a completed Amazon stockout actually cost — lost units and revenue against the pre-stockout run rate, how far sessions and conversion fell, whether they have come back, and how long recovery is taking — then produces a staged plan for restarting ads and rebuilding rank. Use when the user asks what a stockout cost,

Audits how an Amazon Sponsored Ads account is built rather than how it performed last week — ad groups crowded with unrelated products, the same product advertised in a dozen ad groups competing with itself, keywords stacked in every match type inside one ad group, spend sitting outside any portfolio, and selling products with no ad at all —…

Builds the month-end client deck for one Amazon brand — revenue and spend against the previous month and the same month last year, the organic and paid split, category and product mix, what moved and why, and a prioritised plan for next month. Checks that last year's ad data actually exists before drawing a year-on-year comparis

Uses Amazon Marketing Cloud to find which products bring new customers into a brand and which mostly sell to people who already buy it — new-to-brand share by ASIN, month by month, the gateway products worth more ad weight, the retention products that need less, and how much Sponsored Products and Display spend each group gets today.

Builds a mid-month Amazon advertising pacing report — month-to-date spend against the plan, a day-weighted projection to month end, the gap in dollars, which portfolios are capping out before the day is over, which are underdelivering their share, and a reallocation table that moves budget from the second to the first. Use when