BOIM Korea Vendor Directory

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

Annotations already establish read-only/idempotent safety, and the description adds substantial context beyond them: the public data source, that records contain only 상호·업종·행정동 with no detailed address or phone, that ordering is alphabetical rather than rating-based, and that each result carries list_page and ready URLs. For a lookup tool this is unusually rich behavioral disclosure.

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 and the region-required constraint are front-loaded, but the single dense paragraph is heavy with parenthetical asides, and the '[보임 — 한국 업체·입찰 공고 찾기]' prefix references 입찰 공고 (bid notices), which belongs to a sibling tool and mildly muddies the scope.

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 correctly explains what each result returns (list_page, ready URLs) and what fields are absent (no address or phone), which is the key completeness burden. It leaves limit/pagination behavior and the default result count unstated, a modest remaining gap.

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 75% and the schema already documents region narrowing and category examples, which the description largely restates ('지역 필수', 동 단위 축소, 247 subcategories). The undocumented 'limit' parameter is never explained, so the description adds only marginal meaning over 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 specific verb and resource ('전국 영업 중 업체를 ... 찾습니다') plus the full category scope (247 상권업종 소분류, region required), so an agent immediately knows this is a nationwide business-directory lookup, not a bid or vendor-search tool. It explicitly contrasts itself with search_vendors, making sibling disambiguation concrete.

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?

'지역 필수' tells the agent a region must be supplied, and the closing sentence routes to search_vendors when procurement records or owner-registered detail are needed. It does not address the other siblings (find_local_shops, get_vendor, list_categories), so exclusions are incomplete but the primary conditional is clear.

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

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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive, non-open-world), and the description goes well beyond them: it names the data source, discloses what fields exist and do NOT exist (상호·업종·행정동 only, no performance or contact data), states the sort order is by name and explicitly not by rating, and explains the registered_card URL attachment behavior.

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 and dataset provenance are front-loaded, and each sentence carries routing, scope, or behavior information. It is denser than ideal (four-plus clauses in one block) but nothing is padding.

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 4-param read tool with no output schema, the description is nearly complete: it covers what is returned, what is missing, ordering, and the sibling alternative. The only loose end is the 'limit' parameter's behavior, which is left to the schema defaults.

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 75%; work and region already carry examples in the schema, and the description largely restates those categories rather than adding syntax or format detail. The undocumented 'limit' parameter (default 20, max 50) is not explained in either place, and no new meaning is added over the structured fields, so the baseline 3 applies.

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 specific verb+resource (finding neighborhood shops registered in the small-business commercial-information public dataset) filtered by 시·군·구·동, and explicitly scopes the category set (signboards, advertising, printing, copying). It distinguishes itself from siblings by naming search_vendors and the registered_card linkage, so an agent can tell it apart from find_businesses/find_public_bids without opening a schema.

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 an explicit when-to-use trigger ('개인·소규모 일… 맡길 곳을 물을 때 쓰세요') and an explicit when-not with an alternative ('조달 실적이 있는 업체는 search_vendors를 쓰세요'). Routing between siblings is fully specified.

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

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

Annotations already declare readOnly/idempotent/non-destructive/closed-world, so the safety profile is covered. The description adds genuine behavior beyond that: twice-daily refresh from 공공데이터포털, deadline-ascending ordering, and that each result carries an original url and estimated price when available.

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?

Front-loaded with scope and sources, and each subsequent clause (classification rule, refresh cadence, sort order, output fields, intended users) carries distinct information. It is dense and long for a search tool, but nothing is clearly redundant.

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?

No output schema exists, and the description compensates well by describing return fields (url, estimated price, type, origin) and sort order. The `limit` parameter and any pagination behavior remain unexplained, a minor gap for an otherwise complete definition.

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 80%, so the baseline is 3. The description clarifies the meaning of the classification scheme behind `category`/`type` (대분류 mapping, type/origin in results) but adds no syntax or format detail, and the undocumented `limit` parameter (the missing coverage) is left unaddressed.

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 specific verb (찾습니다/find) and resource (마감 전 입찰 공고) plus exact scope: five named sources (나라장터·국방전자조달·LH·수자원·누리장터) and all fields. It also explains how results are classified (group/type/origin), so an agent can immediately tell this apart from the business/shop/vendor siblings.

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 concrete when-to-use scenarios — a company seeking public-institution work, or asking what an institution intends to buy. However it names no alternatives or exclusions, so it does not route the agent away from siblings like find_businesses or search_vendors.

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

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

Annotations already declare readOnly, idempotent, non-destructive, closed-world behavior, so the safety profile is covered. The description adds real value by disclosing the breadth of the payload and, notably, listing a 'data limitations' field, which warns the agent the data has known gaps. It stops short of describing pagination or error behavior, but that is minor here.

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 and prerequisite are front-loaded in one sentence. The long parenthetical field list is dense but earns its place by standing in for the missing output schema. Slightly heavy for a one-parameter tool.

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 usefully enumerates the returned card fields and flags data limitations. Missing only edge-case behavior such as an invalid/missing id, which is a small gap for a simple read tool with full annotation coverage.

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% and the single id parameter is already documented as coming from search_vendors results; the description largely restates that. Baseline 3 applies since the schema does the heavy lifting with only one parameter.

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 specific verb (fetch/retrieve) and resource (the full vendor card) and enumerates the card's contents, so the agent knows exactly what comes back. It explicitly distinguishes itself from search_vendors by naming it as the source of the id, so the two siblings are not confusable.

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 makes the precondition explicit: the id must come from search_vendors results, implying search_vendors runs first. There are no exclusions or named alternatives for when not to use it, but for a single-purpose lookup the context is clear.

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

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

Annotations already establish readOnly/idempotent/non-destructive/openWorld=false, so the safety profile is covered. The description adds real behavioral context beyond that: the 80-item result cap and that items are ranked by vendor count, plus what is returned when `group` is empty.

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?

Two sentences, front-loaded with the identifying context tag and core behavior, then the `group` branch. Every clause carries information (return contents, cap, ordering, prerequisite); nothing is redundant, though the bracketed tag is slightly noisy.

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 carries the return-shape burden and mostly does so: it names the fields returned (category, screen name, procurement name, classification number, vendor counts), the 80 cap, and the empty-`group` fallback. Only exact output formatting is left unspecified.

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% and the single `group` param is documented there, but the description adds meaning the schema lacks: giving `group` yields up to 80 items, while omitting it yields the category list plus the top 5 items per category. That is functional semantics beyond the schema, not a restatement.

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 specific verb and resource: returns procurement major categories plus vendor counts, and with `group` returns item names (screen/procurement/classification) plus vendor counts. This clearly differentiates a taxonomy-listing tool from the vendor/bid search siblings, though the sibling relationship itself is never articulated.

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 explicit sequencing guidance — "전 품목이라 분야를 먼저 고르세요" (pick the category first since it covers all items) — which tells the agent how to drive the tool. It does not name an alternative sibling or state exclusions, so it stops short of a 5.

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

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

Annotations already cover read-only/idempotent/non-destructive safety, but the description goes well beyond them: it discloses the ranking rule (관련도 → AI 준비 구간 → daily shuffle, explicitly not performance or paid ranking), defines the ai_ready field, and explains that results carry a match field showing the evidence (본사/납품 실적/시공 가능 지역). That is meaningful non-obvious behavior for a search tool.

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 and scope are front-loaded, and every clause (ranking rule, ai_ready meaning, region matching, category pointer) carries information an agent needs. It is a dense single paragraph without internal structure or line breaks, which slightly hurts scanability, but there is no obvious filler.

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 6 optional parameters, no output schema and only 83% schema coverage, the description compensates by explaining ordering, the ai_ready and match fields, and region semantics — effectively covering the return behavior. It leaves cert, query and limit to the schema, but nothing critical for correct invocation 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 already 83%, so the baseline is 3, but the description adds real semantics: region matches against three separate sources (본사 소재지, 최근 1년 납품 실적 지역, 등록 시공 지역) with match explaining which fired, and category/group are tied to the 조달청 물품분류 vocabulary and list_categories. This exceeds what the schema alone conveys.

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 names a concrete verb+resource (finding Korean vendor cards built from procurement data) and specifies the underlying sources (종합쇼핑몰 계약 품목 + 최근 1년 조달 실적). It does not clearly distinguish this from siblings find_businesses or find_local_shops, and the bracketed prefix mentioning '입찰 공고 찾기' briefly muddies the line with find_public_bids. Still, an agent can tell this is the procurement-vendor search.

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?

Usage context is implied rather than stated: ordering rules, region-matching semantics and a pointer to list_categories for the category vocabulary are given, but there is no explicit 'use this instead of X when Y' guidance against the five siblings. The agent must infer when this tool is preferable to find_businesses or find_local_shops.

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