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

The annotations already declare readOnlyHint=true, destructiveHint=false and idempotentHint=true, and the description is consistent with them because it explicitly says it 'opens nothing by itself' and only returns instructions. That is a meaningful addition beyond the annotation flags because a name promising a refund would otherwise suggest a mutation. The trailing note about Recurly's REST API describes the vendor, not this tool's behavior, and adds no operational context such as what the returned steps contain or whether they are personalized to the user's account.

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?

The purpose and the handoff to the browser tools are front-loaded and easy to scan. The final sentence about Recurly having a full REST API, including a dated 'checked 14 Oct 2026' provenance note, does not help an agent decide to call this tool or call it correctly, so it is filler that dilutes an otherwise tight two-sentence definition.

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?

With zero parameters and no output schema, the description must carry the workflow, and it does explain that the return value is a step-by-step walkthrough with screens, buttons, and fields. It never explains how the target invoice/account is identified without inputs, nor what the user must have open (Recurly Cloud Browser session, permissions) before the follow-up cotlet_browser_* calls will work, leaving a real gap for an agent planning the task.

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?

The tool takes zero parameters and the schema is fully described, so there is no parameter semantics to explain and the baseline of 4 applies. The one residual ambiguity is how the tool knows which invoice is being refunded when nothing is passed, but that is a scoping question rather than a parameter-documentation defect.

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 states a specific outcome: it returns step-by-step instructions (screens, buttons, fields) for refunding an invoice rather than performing the refund, which usefully corrects the impression left by the name 'recurly_refund_invoice'. It also distinguishes itself from the cotlet_browser_* siblings by naming them as the tools that actually carry out the action. It does not differentiate itself from other recurly_* guidance siblings, but the verb+resource+output form is clear.

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?

There is explicit sequencing guidance: call this to get the walkthrough, then act with cotlet_browser_* in the user's Recurly Cloud Browser, and 'opens nothing by itself' sets the boundary that this tool alone changes nothing. What is missing is when NOT to use it (e.g., when the invoice is already refunded or when the user wants the API instead), and there is no hint about how the target invoice is established given that the tool takes no parameters.

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