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

Annotations already declare readOnlyHint=false and destructiveHint=true, and the description correctly reinforces this rather than contradicting it ('이 호출은 순수 읽기 전용이 아닙니다'). It adds real value by disclosing the side effect that an XBOSS job is created and the result persisted, plus server-side validation rules (attribution month cannot exceed payment month; a tax-base row requires headcount; blank late-payment days are computed by WeTax). It stops short of saying whether a prior save is overwritten, whether the call is idempotent, or what happens on partial validation failure.

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 login precondition is front-loaded and each subsequent sentence carries a distinct rule or side effect. The closing sentence about institution data changes following per-tool procedures is boilerplate that adds little, and the double negative about not being read-only is awkwardly phrased, but overall the text is tight.

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 7-parameter mutation with a required upstream session and a chained dclr_id, the definition covers preconditions, input rules, and the fact that a background job is created. It is silent on the return value (no output schema exists), on where this step sits relative to declarant_save/review/submit, and on whether repeated calls replace or append to previously saved target rows, which is the main thing an agent needs to know before invoking it twice.

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 already 100%, so the baseline is 3; the description earns an extra point by explaining semantics the schema does not, namely the contents of each income_rows entry and the fallback behavior of late_payment_day_count when left blank. The cross-field constraints (attribution <= payment month, headcount required with tax base) are also non-obvious rules absent from 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?

States a concrete verb and object: it saves attribution year-month, payment year-month, and one row per income type (headcount, tax base, withheld local income tax) to the institution. This is far more specific than the Korean title 'save'. It does not, however, distinguish itself from the other *_save siblings (declarant_save, statement_save, adjustment_save), so an agent cannot tell from the description alone which save step it is in the filing flow.

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 prerequisite 'REQUIRES prior wetax__session__login call with same account_link_id' is stated up front and is genuinely useful. Beyond that, there is no guidance on when to use this versus declarant_save, statement_save, review, or submit, nor whether it can be re-run before submission. Usage is only implied by the name and the login precondition.

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