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

With readOnlyHint=false, the bar is lower, but the description still adds substantial context beyond the annotations: it discloses that each call consumes one token, that an XBOSS job is created and its result persisted, and that the call is therefore not a pure read despite querying institution data. It also spells out idempotency-key lifecycle rules and the fact that provider-specific idempotency keys are a different field. It stops short of describing rate limits, error behavior, or what the persisted result contains.

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 description mixes Korean and English in a repetitive way, restates the idempotency rule across several sentences, and includes likely mistranslated phrasing ("anonymous taste calls"). The ordering is also awkward: it opens with a read-only claim and only at the end reverses it with "this call is not purely read-only", forcing the reader to reconcile two statements. Useful content is present but not tightly 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 an 8-parameter tool with no output schema, the description covers the unusual aspects (token cost, job creation, idempotency) but says nothing about what comes back — the payment history records and institution identifiers themselves, or how page_no pagination behaves. An agent knows how to call it safely but not what to expect in the response.

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 88%, so the baseline is 3 and the schema already documents dates, tax_codes, province/municipality codes and page_no. The description meaningfully extends only request_idempotency_key — when to generate it, when to reuse it, and that anonymous calls cannot use it. The remaining seven parameters get no added meaning from the description, and pagination semantics of page_no are left entirely to the schema.

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 names a specific verb and resource set: local tax payment history plus institution identifier lookup, and explicitly excludes payment execution ("납부는 수행하지 않습니다"). This differentiates it from the due-query and refund-query siblings by resource, though it never names those siblings directly. The trailing admission that the call is "not purely read-only" muddies but does not erase the core statement of purpose.

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?

The description supplies a negation (this is not a payment tool) and detailed guidance on when to reuse vs. regenerate the idempotency key, including the case of recovering after a lost response. It does not state when an agent should pick this tool over wetax__tax__local__due__query or wetax__tax__local__refund__query, nor what prerequisites (account_link_id source, session login) are needed.

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