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

Annotations already declare destructiveHint=true and it adds real consequence context: 'Money goes back to the customer and cannot be taken back', the customer is not emailed unless notify_customer is true, and the call is refused before confirmation on three preconditions. It also discloses the confirmation/owner-approval gating, which no annotation expresses.

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?

Front-loaded with what it does and which payment it targets; nearly every sentence carries a distinct rule. It is dense and slightly long, and the paired statements 'By default you confirm each call' and 'Waits for the owner's approval' overlap enough to add mild ambiguity about the gating flow.

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, money-moving tool with no output schema, it covers prerequisites, irreversibility, notification side effects, partial-refund math and approval gating. It does not indicate what a successful call returns or how to verify the refund landed, which is the one gap for a high-stakes mutation.

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 coverage is 100%, so 3 is the baseline, but the description adds semantics the schema lacks: amount includes tax, defaults to the unrefunded remainder, and anything less is a partial refund, with the extra rule that fee refunds must be full. It also makes the notify_customer default consequence explicit.

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?

Names a specific verb and resource set ('Refund a paid booking charge, extra, tip or fee'), states scope ('in full or in part') and the unit of work ('Refunds one payment from list_charges'). This clearly separates it from siblings like cancel_booking (cancels, does not move money) and list_charges (reads).

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?

Explicitly routes the agent: the payment must come from list_charges via payment_log_id plus its booking_id. It states when-not cases (no payment processor, payment not paid, amount too large, gift-card/referral-credit payments must be refunded in the panel) and the constraint that cancellation/reschedule fees are refunded only in full.

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