refund_orderRefund an order back to the original payment source. Refunds the whole order unless item_ids (from get_order) are given. Memberships/packs are terminated on refund unless refund_without_termination is true. [requires scope: money]
| Name | Required | Description | Default |
|---|---|---|---|
| note | No | ||
| reason | No | Cancellation reason (short) | |
| restock | No | ||
| item_ids | No | Partial refund: item_id values from get_order | |
| order_id | Yes | ||
| refund_without_termination | No |
Changes observed during successful MCP inspections. Dates show when Glama detected each change.
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The annotations already indicate destructiveness, and the description adds meaningful behavioral detail: money returns to the original payment source, memberships/packs are terminated on refund unless refund_without_termination is true, and the required scope is [money]. These are exactly the side effects an agent needs before invoking a destructive refund.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three tight sentences: main action, full-vs-partial scope, and the critical membership-termination caveat. No filler, and the most decision-relevant details are front-loaded.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a destructive six-parameter refund tool with no output schema, the description covers the core behavior, the main conditional (item_ids), and the biggest side effect (termination). It doesn't describe response shape or restock/note/reason semantics, but the parameter names are mostly self-explanatory and the required order_id is obvious.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is only 33%, and the description does partially compensate by explaining item_ids ('from get_order') and refund_without_termination's effect on membership/pack termination. Other parameters like note, reason, restock, and order_id rely on their names or schema-only descriptions, so compensation is incomplete.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states a specific action — refund an order to the original payment source — and adds key scoping: the whole order is refunded unless item_ids are provided. It doesn't explicitly name or contrast a sibling tool such as cancel_membership, so it stops short of full sibling differentiation.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
It provides useful conditional usage context: full refund by default, partial refund only when item_ids are supplied, and termination can be suppressed with refund_without_termination. However, it doesn't state when to choose this tool over alternatives like cancel_membership or charge_client, leaving alternative routing implicit.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Most tools have clearly distinct purposes, and the descriptions carefully separate operations like attendance_report, list_classes, and get_class_roster. A few reads overlap conceptually (get_client, get_reservation_history, get_credit_history, get_client_notes all return client-related data), but each has a specific output that an agent can distinguish with reasonable effort.
Tool names overwhelmingly follow a consistent action_object pattern: add_client_note, cancel_class, freeze_membership, refund_order, set_client_tag, unfreeze_membership. The get_/list_ prefix distinction is used predictably, and auth tools (login, logout, whoami) are standard exceptions rather than inconsistent naming.
41 tools is well beyond the 25+ threshold for a heavy tool set. Although each tool appears to serve a real studio-operations need, the sheer number creates a large surface for an agent to navigate and places significant burden on selection accuracy.
The tool set covers the core lifecycle well across clients, classes, memberships, orders, payments, and communications, with no obvious dead ends. Minor gaps exist, such as no update_client profile tool and no class-scheduling creation, but staff workflows can generally be completed.