Skip to main content
Glama

rr_get_recommendation

Read-onlyIdempotent

Fetch one replenishment recommendation by ID with child item provenance, snapshots, linked execution records, and paginated event history.

Instructions

Get one recommendation with child item provenance, snapshots, linked execution records, and event history. The products come in pages: pass next_cursor as cursor until has_more is false; the action's own fields are on every page. The action carries its newest events, oldest first, up to 100 (fewer when they are long); events_total is how many it has and events_truncated is true when older ones were left out, which are on the action's page in the app. Long lists of SKUs or locations on the action are capped at 100 with a _total beside them; the product pages are the complete list.

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": {
      "action_id": "<action_id>"
    },
    "name": "rr_get_recommendation"
  }
}

Input Schema

NameRequiredDescriptionDefault
limitNoRows per page (default 100, max 500).
cursorNonext_cursor from the previous page. Pass it with the same other inputs until has_more is false.
action_idYesRecommendation ID

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv1.3.1

TDQS

A4.4/5.0
Behavior5/5

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

Annotations already declare readOnly/idempotent/non-destructive, and the description adds substantial behavior beyond them: the response is paginated with next_cursor/has_more, fields of the recommendation itself repeat on every page, events arrive oldest-first capped at 100 with events_total and events_truncated flags, and long SKU/location lists are capped at 100 with <name>_total counters. This tells the agent exactly how to interpret a partial result, which annotations alone could not.

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?

Front-loaded with purpose, then grouped by concern (paging, events, capped lists), and every sentence carries operational information. Minor deductions for terminology drift: the tool returns a 'recommendation' but the text repeatedly calls it 'the action', which forces the agent to re-map terms, and the closing sentence partly restates the paging point.

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?

With no output schema, the description carries the full burden of explaining the returned shape, and it does so: nested provenance/snapshots/execution records, event ordering and truncation flags, and totals alongside capped lists. An agent has enough to call the tool, loop the pages correctly, and know when its view is incomplete.

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 baseline would be 3. The description adds real meaning on top: it explains that cursor must be paired with the same other inputs and that iteration continues until has_more is false, and it frames limit implicitly through the page-size model. It stops short of tying limit to the 100/500 caps already in the schema.

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?

States a specific verb and resource ('Get one recommendation') and enumerates the payload it returns: child item provenance, snapshots, linked execution records, and event history. An agent can tell this is the single-record detail reader, distinct from sibling list/detail tools like rr_list_alerts or rr_get_recent_activity.

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

Usage Guidelines3/5

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

Gives concrete how-to-call guidance for pagination ('pass next_cursor as cursor until has_more is false'), which is genuine usage instruction. However, it never says when to choose this over rr_preview_recommendation or when a recommendation ID should come from a list call first, and gives no exclusions or prerequisites.

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