Skip to main content
Glama

Nova Amazon MCP Server

import_supply_chain

Destructive

Import the PRODUCT table of the Supply chain page of ONE account — the same lines as the page's file (2026-10-22): one line per product × warehouse, carrying at most one supplier order (N open orders = N lines, the count and the six numbers repeated): the count at the warehouse / 3PL with its date, an order (units, expected date, PO reference, status), the case pack, the target stock at Amazon and at the warehouse, the transit to Amazon and to the warehouse, the production lead time. The payload changes the lines it names and NOTHING else: a field you omit keeps what is stored, null clears it (a count becomes unknown, an override is removed and the warehouse's number shows again), an order closes with status 'received' / 'cancelled' on its reference; the six numbers are the product's (its 'ALL' line), the same in every warehouse. The count's location and note are not here. On a line straight from the factory (get_supply_chain_products: warehouse.straightFromFactory) there is no warehouse: the count, the transit to the warehouse and the target there are not applied (a null there clears nothing) and the preview returns a notice for the row; put the whole trip, factory to sellable at Amazon, in leadTimeTransitDays. TWO-STEP, never skip the preview: call WITHOUT commit; the server validates (ASINs against the catalog, regions or warehouse names, ranges, dates) and returns per-row warnings, a per-warehouse diff (added lines with their values, changed lines before → after, removed, the warehouse's totals once written) and a confirmationToken; show them to the user and only after explicit approval call again with the SAME account/rows plus commit:true and the token. Up to 2000 rows per call. Requires the OAuth connection and the OWNER role.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "account": "<account>",
      "rows": "<rows>"
    },
    "name": "import_supply_chain"
  }
}

Input Schema

NameRequiredDescriptionDefault
rowsYes
commitNoOmit/false = preview (validate + diff, no write). true = write; requires the confirmationToken from the preview.
reasonNoOne sentence, ALWAYS IN ENGLISH (regardless of the conversation language), on WHY you are making this call — the user question or your analytical goal. Recorded so the seller can see how their data is used.
accountYesThe seller account (seller_partner_id, or an unambiguous account name). One account per call.
confirmationTokenNoThe token the preview returned — proves the user saw exactly this payload.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changed
    • changedInput schema / properties / rows / items / properties / region / description
      Previous value: -"The warehouse's stock region as get_supply_chain_setup lists them ('USA', 'EU_FBA', 'UK', 'Canada'…; 'US', 'EU', 'GB' are accepted), or 'ALL' for the six numbers alone. Give either region or warehouse."New value: +"The warehouse's stock region as get_setup domain=supply_chain lists them ('USA', 'EU_FBA', 'UK', 'Canada'…; 'US', 'EU', 'GB' are accepted), or 'ALL' for the six numbers alone. Give either region or warehouse."
    • changedInput schema / properties / rows / items / properties / warehouse / description
      Previous value: -"The warehouse as get_supply_chain_setup names it (\"UNIS, Buena Park\", or \"UNIS\"): resolved to its region."New value: +"The warehouse as get_setup domain=supply_chain names it (\"UNIS, Buena Park\", or \"UNIS\"): resolved to its region."
  2. First observed

TDQS

A4.5/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=false and destructiveHint=true, and the description goes well beyond them: omitted fields keep stored values, null clears them (count becomes unknown, override removed), orders close via status 'received'/'cancelled', and the preview returns per-row warnings plus a per-warehouse before→after diff and token. The one gap is what a failed or partially rejected write leaves behind after commit, but the destructive surface is otherwise disclosed in detail.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Purpose and the merge-versus-null rule are front-loaded, which is right, but the body is a single dense paragraph of long parenthetical sentences covering line shape, straight-from-factory exceptions and the two-step protocol. Nearly every clause carries real information, yet the lack of any paragraph, bullet or emphasis separation makes the high-value warning (never skip the preview) hard to locate on a fast read.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a high-complexity mutation tool with 2000-row batches, nested row objects and no output schema, the description covers the risky corners: null-versus-omit, order closing, the factory line exception, row limits, auth/role requirements, and what the preview returns. An agent could invoke this correctly without further context.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 80%, so the schema already carries most per-field meaning; the description still adds non-obvious cross-parameter semantics the schema cannot express — that the six numbers belong to the product's 'ALL' line and are identical in every warehouse, and that on a straightFromFactory line the warehouse, warehouse transit and warehouse target are not applied and a null there clears nothing. It does not, however, add much about the account, reason or confirmationToken parameters beyond what their schema descriptions say.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description names a specific verb and resource ("Import the PRODUCT table of the Supply chain page of ONE account") and immediately pins down the unit of work: one line per product × warehouse, at most one supplier order per line. It also draws a boundary against other import tools by stating what is NOT here ("The count's location and note are not here") and points to get_supply_chain_products for source data, so an agent can separate it from import_supplier_orders or import_external_stock.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The workflow is explicit: call without commit for preview, show warnings/diff/token to the user, obtain approval, then call again with identical payload plus commit:true and the token, with "never skip the preview" called out. It also states prerequisites (OAuth connection, OWNER role) and the 2000-row ceiling, leaving essentially nothing to infer about when and how to use it.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.