Japan disability welfare (障害福祉) reference: service fees, add-ons, staffing. In Japanese.
Scored across 2 tools
The two tools have clearly distinct purposes: search_articles discovers relevant excerpts and URLs, while get_article retrieves the full text by slug. There is no overlap that would cause an agent to select the wrong tool.
Both tools follow a consistent verb_noun snake_case pattern (get_article, search_articles). The only minor difference is singular vs plural, which is not confusing.
At two tools, the server is on the sparse side for the typical 3-15 range. However, for a read-only dictionary, search and full-text retrieval are the essential operations, making the count reasonable.
The tool surface fully covers the core dictionary workflow: searching for a topic and retrieving the full article. No obvious missing operations, since the server is read-only and doesn't need create/update/delete.
fukushi.ai の公開記事を slug で指定して全文（Markdown）を取得する。search_articles の抜粋では手順や表の全体が掴めないときに使う。本文末の脚注に、その記事が根拠にした一次情報の出典が示されている。
| Name | Required | Description | Default |
|---|---|---|---|
| slug | Yes | 記事のslug。search_articles の結果に含まれる。例: 「b-type-opening-guide」 |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, so the safety profile is covered. The description adds useful behavioral context: the returned content is Markdown, and the footnotes contain primary-source citations. However, it doesn't mention potential absence of the article, error behavior, or whether the full text is always available. With annotations covering safety, a 3 is appropriate.
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?
Three sentences, each earning its place: the first states the core function, the second gives usage guidance, and the third adds a useful behavioral detail about footnotes. No wasted words.
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 simple read-only tool with one parameter, the description is nearly complete. It covers what the tool does, when to use it, and what the output contains. The only minor gap is not describing behavior when the slug doesn't exist, but the annotations and schema already cover the essential context.
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 schema already documents the slug parameter. The description adds the example 'b-type-opening-guide' and notes that the slug comes from search_articles results, which is helpful but not extensive. Baseline 3 is correct when the schema does the heavy lifting.
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 ('取得する'), a specific resource ('fukushi.ai の公開記事'), and the retrieval key ('slug で指定して全文（Markdown）'). It also explicitly distinguishes itself from the sibling tool search_articles, making the tool's purpose unmistakable.
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 explicitly names the alternative (search_articles) and gives the condition for choosing this tool: when the excerpt from search_articles is insufficient to grasp procedures or tables. This is clear when-to-use guidance with a named alternative.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
障害福祉制度の解説サイト「障害福祉AI辞典」(fukushi.ai) の記事を検索し、該当箇所の抜粋と記事URLを返す。記事は厚生労働省の一次情報（報酬告示・留意事項通知・指定基準省令・サービスコード表）に接地して検証済みで、本文には出典が脚注として示されている。障害者総合支援法にもとづくサービスの制度・報酬・加算・指定基準・実務手順を問われたら、まずこれを使うこと。例:「就労継続支援B型の欠席時対応加算は何単位？」「A型の開業に必要な人員基準は？」「処遇改善加算の算定要件」。扱う範囲は記事がある領域に限られる（就労系と全サービス共通が厚い）。該当が無ければ推測せず、資料に無い旨を利用者に伝えること。
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | 日本語の検索クエリ。サービス種別＋知りたい観点を具体的に並べる。例:「就労継続支援B型 欠席時対応加算 単位数」 | |
| top_k | No | 検索するチャンク数（既定 5、最大 10）。同じ記事の複数箇所が当たると1件にまとめるため、返る記事数はこれより少なくなることがある。 |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations declare readOnlyHint, idempotentHint, and destructiveHint, and the description adds substantial context: it returns excerpts and URLs, merges multiple hits from the same article into one, and has a fallback behavior of stating absence instead of guessing. No contradiction with annotations.
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 description is well-structured: purpose first, then usage examples, then limitations. While somewhat long, every sentence adds value and it is not redundant. Front-loading is effective.
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 adequately explains the return (excerpts and article URLs) and covers edge cases (no results, merging behavior). For a search tool with two params, this is complete and anticipates agent questions.
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%, but the description adds meaningful guidance: query format examples (service type + specific aspect) and clarifies that top_k is a chunk count that may result in fewer articles due to merging. This goes beyond the schema's basic parameter descriptions.
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 clearly states the tool searches articles on a specific site (fukushi.ai), returns excerpts and URLs, and is verified against primary sources. It distinguishes itself from the sibling get_article by focusing on search rather than retrieval, and gives specific examples of queries.
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?
Explicitly instructs when to use this tool: when asked about services' systems, remuneration, add-ons, standards, or practical procedures, and provides concrete example questions. It also states when not to use it (if no article exists, tell the user rather than guessing) and notes scope limitations.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
get_articlesearch_articlesJapan payroll & social insurance. 日本の給与計算・社会保険。47都道府県の料率、源泉所得税、割増賃金、有給、標準報酬月額の改定、最低賃金。根拠の条文つき。
Japan home-visit nursing reimbursement rules, each with its source and that source's standing.
Raw Japanese regulatory data for AI agents: pension, gazette, gBizINFO. x402-metered (USDC).
Japanese law, corporation & statistics data as MCP, normalized to English with source attribution.
Add one secure layer between your agents and this server.