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

Beyond the readOnlyHint/destructiveHint annotations, the description discloses download_url validity (15 minutes), file retention (7 days), the recovery path via xdata__job__result with job_id, the private recovery_receipt semantics for anonymous results (single-result scope, non-recoverable if lost, not transferable), idempotency-key reuse rules, and a per-call cost ('호출마다 토큰 1소진'). That is exactly the class of context annotations cannot carry.

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, but the text is bloated and partly redundant — the result is described twice ('Result: a summary in data ... and the workbook in artifacts[]' then again under 'Files:'), and the receipt/idempotency paragraphs run long with repeated hedging. It is dense but not 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?

With no output schema, the description does the necessary work of describing the return shape (data summary with attribution_ym and employee_count, artifacts[] with signed download_url) plus expiry and recovery. What is missing is error/validation behavior for the required employee fields and any statement of how the readOnlyHint squares with producing a new file.

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?

At 57% schema coverage, the description carries real weight: it names attribution_ym and employees[], and clarifies that deduction/tax amounts are supplied already computed, which is a semantic constraint the schema does not state. It also expands on request_idempotency_key well beyond the schema's one-line pointer. It says nothing about workplace_name/company_name, workplace_address, or business_registration_no, leaving a gap.

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 opening sentence gives a specific verb and resource ('Build a standard payroll register workbook (xlsx)') plus the two driving inputs, which is far more than a restatement of the title. It does not, however, explicitly distinguish itself from the sibling payroll_artifact__payslip_xlsx, so an agent must infer the register-vs-payslip split from the noun alone.

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 parenthetical 'tax/deduction amounts already computed' usefully implies this is a rendering step, not a calculation step, and 'Platform key only — no account_link_id' rules out one path. But there is no explicit statement of when to choose this over payroll_artifact__payslip_xlsx or the payroll_pack__* generators; the bulk of the text is about result retrieval rather than tool selection.

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