An agent that re-checks every shared listing you sell every night, flags where you lost the Buy Box, and prices each loss in dollars — so you fix the listings that cost you most, first.
The agent fetches every listing you sell that has competing sellers on it, with every rival offer and current price.
2
Check who holds the Buy Box
For each listing it records the Buy Box holder across repeated captures — one snapshot lies; a pattern doesn't.
3
Price the loss
Using the listing's estimated sales volume, it computes what each lost Buy Box costs per month: volume × your missing share.
4
Rank and explain
You get a morning report: worst losses first, each with the winning price, your gap, and the likely cause (price, stock, fulfillment).
5
Hand off the fix
Approved fixes go to your repricer or your team as concrete price targets — the audit agent diagnoses, it doesn't guess at prices.
What it needs
Inputs & tools
Your catalogMarketplace seller ID
Offer dataFreshCrawl (live data)
Sales estimatesEstimation engine
Buy Box historyRepeated captures
DeliveryEmail / Slack, 6:30 AM
Your time≈ 15 min/day
Worked example: a toy seller priced $14.94 against a $9.97 Buy Box on one card-game listing, holding the Buy Box in 0 of 8 captures — ≈ $585/month walking out the door on a single listing.
Run this agent yourself
⚡
1 · Get a FreshCrawl subscription
The agent runs on live marketplace data. FreshCrawl plans start free (1,000 credits), paid plans from $29/mo — every plan includes MCP access and an API key. freshcrawl.io →
🔗
2 · Link the MCP to your LLM
In Claude or ChatGPT, add a custom FreshCrawl MCP connector — server crawl.onewally.com/mcp — and paste your FreshCrawl API key. Your LLM can now pull live offers, Buy Box history, and sales estimates itself.
📋
3 · Paste the prompt
Copy the prompt below into your LLM and name your store. The agent runs the full audit against the live data it can now reach — no code, no exports.
Copy the prompt
Starter prompt
Step 3 · Paste this into your LLM
# Buy Box Loss Audit
Use the FreshCrawl MCP connected to this chat for all live marketplace data.
MY STORE (fill in once):
- Store name: [your store name]
- Seller ID or storefront URL: [paste it — used to find your catalog]
- Marketplace: [Walmart US / Amazon US]
First, look up my store in the FreshCrawl MCP, 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: nightly. Report in my inbox by 6:30 AM.
- If your LLM app supports scheduled tasks, set it there; otherwise re-paste the prompt each evening.
For each of my shared listings, find every competing offer and who
currently holds the Buy Box. Repeat the check across the day and
keep only listings where I hold the Buy Box less than half the time.
For each, estimate monthly lost revenue = listing monthly revenue
× (1 − my Buy Box share). Return a ranked table: listing, my price,
winning price, my share, est. $ lost/mo, likely cause. Top 20 only.
EXECUTION — READ CAREFULLY:
You produce the audit only — you have no price-writing access.
Price fixes are executed through Flashpricer (it applies targets
inside my min/max) or by me in Seller Center from your ranked list.
The prompt gets you the shape. The reliable version needs repeated captures and sales estimates — that's the data layer this agent runs on inside Seller Intelligence.
Watch a live example
Load it in your LLM
Same prompt, your assistant
Claude: Settings → Connectors → Add custom connector → paste crawl.onewally.com/mcp and your FreshCrawl API key → open a chat, make sure the connector is on, paste the prompt.
ChatGPT: Settings → Apps & connectors (developer mode) → Add MCP server → paste crawl.onewally.com/mcp and your FreshCrawl API key → new chat, attach the connector, paste the prompt.
Gemini: In Gemini (AI Studio / Gemini CLI) add an MCP server → URL crawl.onewally.com/mcp, header key = your FreshCrawl API key → start a session and paste the prompt.
Grok: In Grok's agent/API console, register an MCP tool server → crawl.onewally.com/mcp + your FreshCrawl API key → paste the prompt as the task. (Grokbot runners can load it as a standing instruction.)
Muse: Connect FreshCrawl once in Muse and just ask — Muse already works this way for the team behind this site; every agent in this directory was documented from live runs like the one below.
Whichever you use, the mechanics are identical: connect FreshCrawl once, paste the prompt, the agent pulls live data itself. Here's an actual run:
▶ Live run · Buy Box Loss Audit — Toywiz on UNO Show 'em No Mercy
Use the FreshCrawl MCP connected to this chat for all live marketplace data.
Run a Buy Box Loss Audit for Toywiz on the UNO Show 'em No Mercy listing (item 1589091315): who holds the Buy Box, what's my share, and what is it costing me per month?
FreshCrawl MCP · get offers item 158909131533 offers found
FreshCrawl MCP · buy box history last 8 capturesWalmart.com 8 / 8
Finds below-MAP listings across the web daily and builds the enforcement evidence pack.
Questions sellers ask
How does the agent know which listings are mine?
From the MY STORE block at the top of the prompt: you give it your store name plus your seller ID or storefront URL once. The agent looks your store up in the FreshCrawl MCP, pulls your catalog, and audits only your listings. If it can't identify your store, the prompt tells it to stop and ask for your storefront URL instead of guessing.
How does an agent know who holds the Buy Box?
It captures the listing repeatedly (consumer and business views) and records the winning offer each time. A seller's share of captures is their Buy Box ownership — the number that decides who actually gets the sales.
How is "lost revenue" calculated?
Estimated monthly revenue for the listing × (1 − your Buy Box share). It's an estimate — the Buy Box holder captures most, not literally all, sales — and it's labeled as one everywhere it appears.
Can the agent fix the prices too?
Diagnosis and repricing are separate jobs. The audit hands ranked targets to a repricing agent (or your team) that moves prices inside guardrails you set — that's the Repricing Watchdog pattern.
Which marketplaces does this work on?
Any marketplace with a featured-offer mechanic: Walmart and Amazon today; the same pattern applies anywhere one offer wins the default purchase.
Deploy this agent on your store.
Seller Intelligence runs the Buy Box Loss Audit on live FreshCrawl data — start with the free report for your store, every losing listing and dollar gap included.