Your AI, on your account. All of it, at once.
This is SellerDragon's MCP interface – the same seller account the dashboard and Sheets read, exposed to the AI client you already use. Your model does the reasoning; SellerDragon supplies the data, the history and the action tools. One question sweeps every campaign, every product, every keyword: what's bleeding, what moved, what changed after last week's calls. Wide decisions land as one staged batch, on your go.
This is an operating layer – staged writes, verified submissions, and a memory that outlives the conversation.
What your AI can reach
The whole account, one question away.
The same isolated data the dashboard and Sheets read – exposed to your AI as real tools. And an honest split: for one-bid-at-a-time detail work the dashboard is faster. The MCP earns its keep at width – account-wide sweeps, batch rulings, and the week-later read.
The superpower is time. Seller Central shows you today – a snapshot here, a report there. SellerDragon keeps your history and hands it to your AI whole: ranks, spend, stock and search lined up on one timeline. "What changed, when, and what moved with it" is the question Amazon can't answer cleanly. Here it's one sentence. And a week after a wide batch of changes, the same timeline closes the loop – ranks, search queries and spend lined up to show how organic sales moved after the changes.
Performance
- Sales trends by ASIN, SKU or family
- Ads summaries and campaign dossiers
- TACoS at account and product-group level
- Every answer anchored to marketplace time, not server time
Keywords & search
- Organic ranks and rank movement
- SQP funnels and search-term performance
- Keyword opportunities and market share
- Over-time correlation – ranks vs bids vs stock events
Inventory
- Stock cover at the current run rate
- Levels, locations and inbound
- Stranded and problem stock flags
Listings
- Full listing detail, issues and schemas
- Staged edits, creates and deletes
- Family-wide batches, one approval
PPC actions
- Staged bid changes and budgets
- Negatives from search terms – incl. the store-wide zero-sale leak list
- Campaign creates, verified after submit
Memory
- Notes with action dates – "recheck in 7 days", asked back as "what's due today?"
- Playbooks – standing instructions your AI follows
- Change history: who did what, when
Reads, memory and listing writes are on every plan. Staged PPC writes start at Growth.
Memory · notes & playbooks
It remembers, so you don't have to.
SellerDragon's MCP carries two kinds of memory – the account's, and yours.
Notes with action dates live on campaigns, ASINs and keywords. Stage "recheck this bid in 7 days" and it surfaces when it's due – open the morning by asking "what needs doing today?" and the answer is your own follow-ups, on time. (Your AI answers when asked – it doesn't message you on its own.)
Playbooks hold your doctrine – how you optimize campaigns, how listings get written, how inventory is watched. Your AI reads them before proposing anything, so suggestions follow your doctrine, not generic best practices. And the context rides along by itself: campaign and keyword dossiers arrive with your relevant notes and playbook already attached.
And the memory lives in SellerDragon, not in the chat window. Switch AI clients, start a fresh session, open the dashboard – same notes, same playbooks, same due list. It's the account's memory, not the conversation's.
The control model
Ask. Stage. Commit.
Approve every change yourself, or let your AI commit on its own – how much rope it gets is your setting, not ours. Either way, writes are never direct: every change passes the same three gates, inside customizable guardrails no AI can bypass.
Ask
Your AI reads the account: sales, ranks, search terms, stock cover, listings, campaign performance. Reads are instant and read-only – ask anything, nothing changes.
Stage
Writes are proposals. Bid changes, negatives, budgets, listing edits land in a staging queue with the exact before/after – nothing touches Amazon at stage time, and every proposal is checked against your guardrails.
Commit
The staged change ships on a go – yours by hand, or your AI's under standing orders you've granted. Either way it goes through Amazon's official APIs and is verified afterwards, with every stage, submit and verification recorded in the audit trail.
How you and your AI split the work is your call – your playbook is the working agreement: require a go on every change, or let it run inside the guardrails you set.
Guardrails
Limits the AI can't cross. You set them.
Every account carries a set of hard limits – max price change per request, max PPC bid, max campaign budget, banned listing terms, whether listing deletes are allowed at all.
They're enforced in the API, not in the prompt: a staged change outside the limits is simply refused. No AI can bypass them – and by design, no AI can edit them. Only the account owner changes a guardrail.
Set them tight and let your AI run, or loosen them as trust grows – they're your dials.
Included everywhere
MCP is on every plan.
Not an add-on, not an upsell. Every plan includes reads, deep analysis, memory and staged listing writes. Staged PPC writes start at Growth, and monthly call volume scales with plan.
Getting connected
Bring the AI you already use.
Connect Amazon
Create your SellerDragon account and authorize your seller account – Amazon's official SP-API and Ads API, scopes you approve, revoke any time.
Add the connector
Add SellerDragon as a connector in your AI client and sign in. Nothing to install – it works wherever your AI lives.
Ask
Open with "What's due today?" or "How were sales this week?" – your account answers.
FAQ
Asked and answered.
What is an Amazon seller MCP server?
MCP (Model Context Protocol) is the open standard AI clients use to work with outside tools. SellerDragon's MCP server exposes your Amazon seller account to the AI you already use – Claude, ChatGPT or any MCP client – as real tools: sales, ranks, stock, search data and staged changes. Your AI does the reasoning; SellerDragon supplies the data, the history and the action tools.
Can I connect Claude or ChatGPT to my Amazon seller account?
Yes. Connect Claude, ChatGPT or any MCP-capable client to SellerDragon and your AI can read sales, ranks, inventory and PPC performance, and stage listing or PPC changes that ship on your terms. You bring your own AI subscription; MCP access is included in every plan.
Can the AI change my account on its own?
Only if you let it. Every write is staged with the exact before and after, and commits on your go – or under standing orders you grant, inside hard guardrails (max bid, max budget, price-change limits) enforced in the API. No AI can bypass or edit the guardrails, and every change lands in the audit trail.
What happens if the AI suggests something wrong?
A bad suggestion is a proposal you discard – nothing touches Amazon until a staged change is committed, and even then the guardrails cap what any single change can do. The dashboard shows the exact before and after for verification.
Stop copy-pasting your account into a chat window.
Connect your AI to the real numbers, with a hand brake on every write.
Same isolated per-account data as every other surface. Buyer PII is discarded before it ever reaches a database.