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

Annotations only say read-only/non-destructive, and the description goes far beyond that: per-call token cost, partial/computed_error failure semantics, the two failure envelopes, 15-minute download links and 7-day file retention, refresh via xdata__job__result, and the anonymous recovery_receipt contract (do not publish, cannot be repaired). These are exactly the operational traits an agent cannot get from structured fields.

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?

A single dense run-on paragraph mixing English and Korean, where the most important caveat (these are NOT the employee deduction amounts) is buried mid-sentence and the receipt/artifact plumbing occupies as much space as the conversion rule. The content mostly earns its place, but the packing and ordering make it hard to scan.

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 11-parameter composite with no output schema and only safety annotations, the description supplies the return-value shape (artifacts[], computed, partial, computed_error), the failure envelopes, and the recovery path, which is close to sufficient. What remains missing is any indication of relative cost/latency across the three leg fetches or a fallback when only one provider link is missing.

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 100%, so the baseline is 3; the description adds real semantics on top by explaining that a 고지 reissued as multiple NHIS 작성회차 resolves to the newest revision, and that row keys exist to disambiguate ties or reach an older notice. It also ties output_formats to artifact behavior. It does not explain how the required notice_year_month / installment_round / billing_year / billing_month map onto the three provider legs, which would have earned a 5.

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?

Opens with a specific verb+resource: combines NHIS, NPS and COMWEL billing notice steps and returns computed premium totals. The second sentence sharply delimits the output ('institution-notice totals billed to the workplace, including BOTH employee and employer portions ... NOT the employee payroll-deduction amounts'), which prevents the most likely misuse. It does not, however, distinguish itself from the sibling payroll_pack__premium_summary_v1__generate or employment_insurance_summary_v1__generate, which an agent would need to choose between.

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?

Gives prerequisites ('Requires prior session login for 건강보험·국민연금·고용산재') and an explicit negative rule ('No account_link_id is needed here'), plus what to do when computed is null. But it never says when to use this pack instead of the closely-named sibling packs, so alternative selection is left to inference.

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