Propose: refund

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

Annotations declare destructiveHint=true and openWorldHint=true, which on their own would suggest immediate money movement. The description usefully corrects that impression by disclosing the deferred, human-gated execution model and the apply_proposal dependency, which is real value beyond the annotations. It does not cover what happens on rejection, whether the proposal can be edited, or what api_refund triggers downstream.

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?

Two short sentences with no filler, and the core action is front-loaded before the safety caveat. Every clause carries information the agent needs.

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?

There is no output schema, so the description should say something about what a successful call returns (a proposal identifier?) and whether amount is validated against the order total. It covers the safety model well but leaves the parameter ambiguity and the return contract unaddressed, which is a meaningful gap for a five-parameter money-related tool.

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 40% across five parameters, and the description adds nothing beyond the parenthetical '(partial)' hinting that amount may be below the order total. Most importantly, the two similarly named required string parameters, 'reason' and 'refund_reason', are left completely unexplained, so an agent has no way to know how they differ or what content belongs in each.

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 uses a specific verb and resource ('Propose a (partial) refund on an order'), which separates it cleanly from read-oriented siblings like get_order and from the other propose_* tools by resource. It stops short of explicitly naming a contrasting sibling, but the resource name alone makes the choice unambiguous.

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?

Mentioning that money moves only after human approval and apply_proposal runs implicitly tells the agent this is step one of a two-phase workflow and names the follow-up tool. However, it never states when to prefer this over alternatives, whether a direct refund path exists, or what conditions make a refund appropriate, so the guidance is implied rather than explicit.

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