An agent that reconciles your Amazon FBA reports line by line — inbound shipments, warehouse losses, return refunds, fee tiers — and turns every discrepancy into a documented case, ready to open. Amazon keeps what sellers don't claim; this agent claims it.
Inventory ledger, inbound shipment reports, returns and refunds, fee transactions, removal orders — the full paper trail, on schedule.
2
Reconcile every unit
Shipped vs received, sold vs adjusted, returned vs restocked. Any unit Amazon's reports can't account for becomes a candidate claim.
3
Audit the fees too
FBA fulfillment fees are tiered by size and weight: wrong dimensions mean a wrong tier on every single order. Storage and long-term storage charges get the same treatment.
4
Build the case file
Each claim gets its evidence pack: report lines, quantities, dates, the policy it falls under, and the amount — written the way Seller Support expects to read it.
5
Track to the money
Opened cases are followed to their outcome. Short-paid reimbursements get re-filed; reversed reimbursements get challenged; expiring claim windows get flagged before they close.
What it needs
Inputs & tools
Your reportsSeller Central / SP-API exports
Fee checkYour measured dims + fee reports
Claim floorOpen cases ≥ $25 (your call)
CadenceWeekly + monthly sweep
DeliveryClaim file, Mondays 6:30 AM
Your time≈ 15 min/week approving
The standard find: an inbound shipment of 500 units received as 486. Fourteen units sit un-reconciled for months because nobody had time to chase a case. The agent's job is making sure 'nobody had time' stops costing money.
Run this agent yourself
📑
1 · Bring your account reports
This agent's source of truth is your own account data — the inventory, shipment, returns and fee reports from Seller Central (or Walmart Seller Center). Export them — or integrate with the marketplace APIs directly (Amazon SP-API, Walmart Marketplace API): the LLM you're using can write that integration with you, and then the reports pull themselves on schedule. No public crawl can see them; they're behind your login, as they should be.
🔗
2 · Optional: link the FreshCrawl MCP
Add the FreshCrawl MCP connector — server crawl.onewally.com/mcp — and the agent can cross-check what your public listing pages state (dimensions, weight, current price) against your reports. Supporting evidence only: disputes rest on your measured actuals and your report lines.
📋
3 · Paste the prompt
Copy the prompt below, attach your reports, name your store. The agent reconciles everything and hands you the claim file — no code.
Copy the prompt
Starter prompt
Step 3 · Paste this into your LLM
# FBA Reimbursement Hunter
MY STORE (fill in once):
- Store name: [your store name]
- Seller ID or storefront URL: [paste it — the agent uses it to find your catalog]
- Marketplace: [Walmart US / Amazon US]
First, use the FreshCrawl MCP to look up my store, pull my catalog,
and work only on my listings. If you can't identify my store from
what I gave you, ask me for my storefront URL before anything else.
RUN SCHEDULE (set this where you run the prompt):
- Run this prompt: weekly inbound check Mondays 6:30 AM, plus a monthly deep sweep on the 1st.
- If your LLM app supports scheduled/recurring tasks, set it there; otherwise re-paste the prompt on that cadence.
DATA SOURCES: my own account reports are the source of truth. Feed them to you two ways: (a) I attach the exports, or (b) we integrate with the Amazon SP-API (Reports API) directly — if no integration exists yet, write it with me first, then pull the reports yourself on schedule. Use the FreshCrawl MCP, if connected, only to cross-check what my public listing pages state (dimensions, weight, current price) — claims rest on my measured actuals and my report lines, never on the crawl alone.
Reconcile my Amazon FBA reports (inventory ledger, inbound shipments, returns/refunds, fee transactions, removals — I'll attach the exports or connect my account) and check all ten leak categories:
1. Inbound shipments: units shipped vs received.
2. Inventory lost in Amazon warehouses and never located.
3. Inventory damaged by Amazon (warehouse damage).
4. Customer refunds where the item never came back.
5. Refund errors: wrong amounts, refunds outside the window, double refunds.
6. Removal/disposal orders with missing or damaged units.
7. FBA fee overcharges: wrong size/weight tier (my measured dims are the evidence; cross-check the listing's stated dims via FreshCrawl where available).
8. Storage and long-term storage fee errors.
9. Unexplained inventory adjustments and wrong-ASIN receipts.
10. Reimbursements short-paid, or paid and then reversed.
For every discrepancy: claim amount, evidence (report + line), the filing window it is in, and a drafted case in Seller Support's format. Sort the claim file by dollars, biggest first.
EXECUTION — READ CAREFULLY:
You find and document claims and draft cases. Never invent a quantity or a dollar figure — every claim must cite the exact report and line it came from. Where I have connected my account so you can open cases, open only claims of $25 or more; draft everything else for my approval. Amazon decides every case: write 'claimable', never 'owed', and track each case to its outcome so reversals and short-pays get re-filed.
The prompt gets you the shape. The reliable version runs on live data through the MCP — that's steps 1 and 2 above.
Clusters return comments to find the few fixable causes behind most returns.
Questions sellers ask
What does it actually check?
Ten categories: inbound shortages, warehouse-lost and warehouse-damaged units, refunds without returns, refund errors, removal discrepancies, FBA fee tier overcharges from wrong dimensions, storage fee errors, unexplained adjustments, and short-paid or reversed reimbursements.
Does it open the cases itself?
It builds every case file complete with evidence. If you connect your account so it can file, it opens claims above the floor you set ($25 default) and drafts the rest for your approval — Amazon still decides each case, which is why the agent says 'claimable', never 'owed'.
Why does this need an agent at all?
Because the leaks are individually small and collectively large — commonly reported at 1–3% of FBA revenue — and each claim type sits in its own report with its own filing window. Humans sample; the agent reconciles everything, every week.
Can it recover old discrepancies?
It works backward through your reports as far as each claim type's filing window allows — windows have tightened in recent years, which is exactly why the sweep runs weekly instead of whenever someone remembers.
Run this on your store.
Every agent here runs on live FreshCrawl data through the MCP. Start free, connect your LLM, paste the prompt — or get the free Buy Box Loss Report and see the data first.