Always-On Ad Audits Across 15,000 Campaigns with Claude

Here is my search term report for the last [14] days [attach].
Give me, in plain English: the search terms wasting money (spend over $[x] with no orders) and the ones worth scaling (orders with ACoS under [y]%).
For each, the exact change you'd make. Don't make any change until I confirm it.

You have read access to my Amazon Ads and sales data. Don't change anything without my approval.
1. List my current campaigns and the products each one covers. Flag any campaign that mixes different product formats.
2. Propose a structure with one campaign per product format, each with its own keyword list pulled from search terms that already convert for that format.
3. For every change you want to make (new campaign, bid, budget, negative), tell me what and why, and wait for my yes.
4. Each week, report spend, CPC, sales and units by product format against the week before.

Audit my Sponsored Products search terms for the last 60 days. List every search term with spend and zero orders, grouped by campaign, with clicks and spend.
Before you change anything, write out the full change list: campaign, the negative you'd add, match type, and why. Flag any term that converts in other campaigns, since blocking it everywhere could cut sales. Wait for my OK.

I'm attaching two files for the same brand: the Sponsored Products search term report for the last 60 days and a Helium 10 Cerebro export of the brand's organic ranks.
Match them by keyword and give me three lists:
1. Keywords where we already rank organically in the top 10 but still spend heavily on ads (candidates to cut spend).
2. Keywords ranked 11-40 with good conversion in the ads data (candidates to push toward page 1).
3. Keywords with spend and no organic rank at all.
Show spend, sales, ACoS and organic rank for each row.

You are my Amazon data engineer. Write a plain-Python script (standard library only) that:
1. Uses my Amazon Ads API refresh token to pull daily Sponsored Products campaign stats and the full search-term report for the last 30 days, plus the same 30-day window last year.
2. Uses SP-API to pull Sales & Traffic by ASIN, FBA inventory, and customer returns for the same windows.
3. Saves everything as one JSON bundle per day.
Then write a reconciliation check: for a date I choose, print total spend, clicks, orders and sales so I can compare them against my current PPC tool. Do not make any write calls to Amazon.

Find all keywords across all campaigns that have spent more than $50 with zero sales in the last 30 days.