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

This goes well beyond the annotations, disclosing success costs at runtime: '호출마다 토큰 1소진,' that an XBOSS job is created and its result stored, and that the call is 'not purely read-only' even though it reads from the source — which correctly explains the readOnlyHint=false annotation on a lookup tool. The idempotency-key lifecycle and the warning that the provider's own idempotency_key is a different field are useful additions; minor deductions for garbled phrasing ('anonymous taste calls') and the tension between 'generate ... before the first request' and calling the key 'optional.'

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 precondition is correctly front-loaded, but the body is padded with restatement ('Without it, calling again remains a new request,' 'This optional key is not available to anonymous taste calls') and repeats its content across English and Korean segments without adding information. Several sentences could be merged or cut without loss.

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 two-parameter tool with no output schema, the definition covers the prerequisites, the per-call token cost, the side effects, and the idempotency contract, which is the information an agent needs to call it safely. The main gap is that it never indicates what data 'workplace basic info' actually contains, so the agent cannot judge result relevance before invoking it.

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 100%, so the account_link_id semantics (UUID v4, optional, server resolves from API key, per-institution values) are already well documented. The description still earns credit by carrying the substantive meaning of request_idempotency_key, whose schema entry only says 'see the tool description' — the retry-versus-new-execution distinction is defined only here.

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 states a specific verb and resource: 'Get workplace main basic info from NPS,' which matches the title '사업장 기본정보 조회 (국민연금)'. Naming the NPS source distinguishes it from the NHIS and opendata siblings in the catalog, though it never explicitly contrasts behavior with those alternatives.

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?

'REQUIRES prior nps_edi__auth__npki_session__login call with same account_link_id' is a hard, actionable precondition that tells the agent exactly when this tool can be invoked. It also gives clear when-to-reuse guidance for the idempotency key (only to recover a lost response, new key for new execution), but offers no guidance on when to prefer a sibling lookup tool instead.

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