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

Annotations already establish readOnlyHint=true, so the safety profile is covered. The description adds genuinely non-obvious behavior on top: the auth dependency, that each call consumes one token (호출마다 토큰 1소진), and precise retry semantics for request_idempotency_key, including that omitting it makes a repeat call a new request. Return shape and pagination behavior are not disclosed, and the "anonymous taste calls" phrasing is confusing.

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 hard prerequisite is correctly front-loaded, but the idempotency explanation sprawls over four sentences with redundancy ("Without it, calling again remains a new request" restates the preceding clause) and an odd aside about anonymous calls. The trailing Korean token-cost note is important but buried at the end.

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 10 parameters and no output schema, the description covers the gating concerns (auth, idempotency, token cost) but says nothing about the date-range/template_code filters or pagination defaults that drive correct invocation, nor what a result row contains. Adequate for a read-only search but not complete.

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, and the description earns a bump by supplying the only real semantics for request_idempotency_key (the schema merely says "see the tool description") — namely when to reuse it for lost-response recovery versus minting a new key. It also usefully clarifies that the provider-specific idempotency_key is a distinct field, preventing a mix-up with account_link_id conventions.

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?

"Search pension premium notices" states a specific verb and resource, so the core operation is unambiguous. However, it does nothing to distinguish this from close siblings such as nps_edi__b1__pension_notices__select, nps_edi__b1__pension_notification_documents__search, or nps_edi__b1__lookups__workplace_premium_notices__search, leaving the agent to infer the boundary.

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 clearly states a prerequisite (a prior nps_edi__auth__npki_session__login with the same account_link_id) and explains when to reuse versus regenerate the idempotency key. But it gives no guidance on when to choose this tool over the select/documents-search siblings, which is the more consequential selection decision here.

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