Skip to main content
Glama
Amazon Ads MCP Server

Amazon Ads MCP Server

by get-mcp-ads

amazon ads create report

amazon_ads_create_report
Read-onlyIdempotent

Request asynchronous Amazon Ads performance reports for Sponsored Products, Brands, or Display, choosing the attribution window explicitly. Get a jobToken back, then fetch results later.

Instructions

Request an asynchronous Amazon performance report: Sponsored Products campaigns, ad groups, placements, search terms or advertised products; Brands campaigns, ads or search terms; Display campaigns or products. Brands legacy campaign reports exclude multi-ad-group campaigns. Does not change advertising or budgets. Choose the attribution window explicitly, keep jobToken and call amazon_ads_get_report later. No polling loop, no cross-platform attribution and no automatic sums.

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": {
      "accountId": "<accountId>",
      "startDate": "<startDate>",
      "endDate": "<endDate>",
      "attributionDays": "<attributionDays>"
    },
    "name": "amazon_ads_create_report"
  }
}

Input Schema

NameRequiredDescriptionDefault
endDateYesLast day of the report, YYYY-MM-DD, at most 30 days after startDate.
timeUnitNoDaily rows, or one aggregate row per grouping over the requested range.DAILY
accountIdYesAdvertising profile ID listed in AMAZON_ADS_PROFILE_IDS, such as 1234567890. The region-qualified form used by hosted getmcpads, such as EU:1234567890, is accepted when its region matches AMAZON_ADS_REGION.
startDateYesFirst day of the report, YYYY-MM-DD, in the profile timezone.
reportTypeNoNative report type. sp* reports cover Sponsored Products, sb* Sponsored Brands and sd* Sponsored Display.spCampaigns
attributionDaysYesNative click attribution window in days for purchases and sales. Brands and Display accept 14 only. Never compare different windows as equivalent.

Output Schema

NameRequiredDescriptionDefault
resultYesOriginal tool result: parsed JSON when the text is JSON, otherwise the text or multiple MCP content blocks. Provider fields depend on the selected query.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed4 schema fields changedv1.0.1
    • removedInput schema / properties / endDate / $ref
      Removed value: -"#/properties/startDate"
    • addedInput schema / properties / endDate / pattern
      Added value: +"^\\d{4}-\\d{2}-\\d{2}$"
    • addedInput schema / properties / endDate / type
      Added value: +"string"
    • changedOutput schema / properties / result / anyOf
      Previous value: -[
      -  {
      -    "additionalProperties": true,
      -    "properties": {},
      -    "type": "object"
      -  },
      -  {
      -    "items": {},
      -    "type": "array"
      -  },
      -  {
      -    "type": "string"
      -  },
      -  {
      -    "type": "number"
      -  },
      -  {
      -    "type": "boolean"
      -  },
      -  {
      -    "type": "null"
      -  }
      -]New value: +[
      +  {
      +    "additionalProperties": {},
      +    "properties": {},
      +    "type": "object"
      +  },
      +  {
      +    "items": {},
      +    "type": "array"
      +  },
      +  {
      +    "type": "string"
      +  },
      +  {
      +    "type": "number"
      +  },
      +  {
      +    "type": "boolean"
      +  },
      +  {
      +    "type": "null"
      +  }
      +]
  2. First observedv1.0.0

TDQS

A4.5/5.0
Behavior5/5

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

Goes well beyond the readOnlyHint/idempotentHint annotations: it discloses asynchronous execution, the jobToken persistence requirement, that no polling loop is provided, that attribution windows are not comparable, and that no automatic sums or cross-platform attribution occur. It also reinforces the non-mutating nature with 'Does not change advertising or budgets'. This is exactly the extra context annotations cannot carry.

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

Conciseness4/5

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

Four sentences, front-loaded with the verb and resource, followed by scope, constraints, and the follow-up call. Dense but every clause carries information; the sentence listing every report type is long but is doing necessary disambiguation work.

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

Completeness4/5

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

With an output schema present, return values need not be explained, and the async lifecycle is fully covered (create here, retrieve via get_report). Minor gaps remain: no mention of the 30-day max range or the region-qualified accountId form, though both are handled in the schema descriptions.

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 100%, so the schema already documents every parameter, and the description correctly does not repeat it. It adds interpretive guidance on the one ambiguous choice — 'Choose the attribution window explicitly. Never compare different windows as equivalent' — which is semantics the enum alone does not convey.

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?

Names a specific verb and resource ('Request an asynchronous Amazon performance report') and enumerates exactly which report families are covered (SP campaigns/ad groups/placements/search terms/products, SB campaigns/ads/search terms, SD campaigns/products). The async nature plus the jobToken handoff cleanly separates it from the get_* entity-listing siblings.

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

Usage Guidelines4/5

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

Explicit follow-up path is given ('keep jobToken and call amazon_ads_get_report later'), along with exclusions ('No polling loop', 'no cross-platform attribution', 'Brands legacy campaign reports exclude multi-ad-group campaigns') and the instruction to choose an attribution window explicitly. It lacks an explicit statement of when to prefer this over the many get_campaigns/get_targets siblings, but the performance-vs-configuration split is strongly implied.

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