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

Annotations already declare readOnly/idempotent/non-destructive, and the description adds genuinely new behavior: unknown codes return found: false with nearby codes and never a fabricated rate, the rate is keyed to the declaration export date rather than the invoice date, lookup.in_force_on names the reference day, and the notes must be relayed to the user with the non-official-source caveat.

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?

Purpose is front-loaded and the sentences are individually tight, but the entire content is duplicated in Chinese and English, roughly doubling the length with no new information. The repeated 'never guess today's date' instruction also appears three times across the description and schema.

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 no output schema, the description carries the return contract itself: every period (refund rate, levy rate 征税率, from, to), the in-force period, in_force_on, and the found:false/nearby-codes failure shape. Combined with the annotations, an agent has everything needed to call and interpret the 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 100%, and the schema entries for code and export_date already carry the same format and 'not the invoice date / Beijing calendar' semantics the description repeats. Per the rubric, repeating structured data earns no extra credit, so the baseline 3 applies.

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 precise verb+resource: look up the export VAT refund rate of ONE customs/HS code in the 出口退税率文库, with accepted formats (8/10 digits, dots/spaces) and an example. The word 'ONE' implicitly scopes it as a single-code lookup. It does not name or contrast with the sibling 'tuishui', which is the most plausible overlap, so it falls 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 gives explicit trigger phrasing ('帮我查一下海关编码 … 的出口退税率'), states exactly when to supply export_date (the customs declaration export date) versus when to omit it ('现在'/'今天'), and warns never to guess today's date. It also says to state in the reply which day was used. This is about as complete a when/when-not/behavioral-routing set as a lookup tool can carry.

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