オンライン学習講座　社員教育にも使える

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

Annotations already declare readOnlyHint=true, destructiveHint=false, and openWorldHint=false, so the safety profile is covered. The description adds that it returns the *full* body text and that accuracy-critical facts should be checked here, but it does not mention truncation, pagination, or any auth/rate considerations beyond the 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?

Three sentences, front-loaded with the core action (returns full body text) followed by the id usage and the accuracy recommendation. Nothing is redundant, though the accuracy sentence is somewhat verbose.

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 an output schema present and 100% parameter coverage, the description need not explain return values, and it appropriately focuses on scope and the id argument. It is complete enough to call correctly, though it omits any distinction from the near-identical `get_page` sibling.

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% and the single `id` parameter is fully documented in the schema (search-result id, URL, or path, Japanese allowed). The description largely restates the same accepted forms without adding syntax or formatting detail beyond the schema, 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?

The description uses a specific verb+resource in Japanese: it returns the full body text of a page from the named site (online.instantgakushu.jp). An agent can tell it retrieves page content, but it does not differentiate itself from the sibling `get_page`, which appears to overlap, so it falls short of a 5.

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?

It gives an explicit usage condition: verify prices, dates, conditions, and terms against this body text before answering. However, it never names an alternative (e.g., when to prefer `get_page` or `search_pages`) or states when not to use it, so it lacks the routing clarity 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 declare readOnlyHint=true, destructiveHint=false and a closed world, so the safety profile is covered. The description adds valuable non-annotation context: this is a deprecated alias retained for backward compatibility and its output shape differs from fetch.

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?

Short and front-loaded with the most decision-relevant fact (legacy feature). The parenthetical domain and the duplicated deprecation messaging are slightly redundant but not wasteful.

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 single-parameter, read-only legacy tool with no output schema, the description says what it returns, how it differs from fetch, and why it exists. Nothing critical for a 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?

Only one parameter with 100% schema coverage, including the note that Japanese URLs/paths are acceptable. The description adds nothing about the url argument, 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 ('ページの本文を返す') and explicitly distinguishes itself from the sibling fetch by noting the return format differs. The legacy framing ('以前の名前の機能') is clear, though the identity of the tool is somewhat overshadowed by the deprecation notice.

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 names the alternative ('新しく使うときは fetch を使ってください') and the condition that selects it. It also explains why it still exists ('すでにつないでいる人のために残しています'), which tells the agent when NOT to pick it.

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 readOnlyHint=true, destructiveHint=false and openWorldHint=false, so safety is covered structurally. The description adds real behavioral value on top: page size (50), the limit ceiling (200) and offset-based continuation, plus what the payload contains (titles and URLs). It stops short of describing result ordering or empty-result 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?

Three sentences ordered as content → pagination → when-to-use, with no filler. The pagination constraints are front-loaded in the middle sentence where an agent will look for them.

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 accepts the burden of describing the return shape and does so ('タイトルと URL'), and pagination is fully explained. Missing only secondary details such as ordering or whether a total count is available, which are minor for a list tool whose annotations already cover safety.

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 0%, so the description must carry the semantics itself, and it does: limit's default of 50 and maximum of 200, and offset as the continuation cursor. Only the minimum bound (1) and the offset step size are left to the schema, so the compensation is substantial but not exhaustive.

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 ('読み込んだページの一覧を、タイトルと URL で返します') and pins the scope to one domain (online.instantgakushu.jp), including the returned fields. It distinguishes itself implicitly from the search siblings ('探しても見つからないときに使います') but never names search_pages, get_page or site_overview, so an agent must 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?

Gives two concrete trigger conditions: browsing what pages exist, and falling back when a search finds nothing. That is clear contextual guidance with no exclusions stated and no sibling named, which keeps it just below the top mark.

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 readOnlyHint=true, openWorldHint=false and destructiveHint=false, so the safety profile is covered. The description adds genuinely useful behavioral context beyond that: the corpus boundary (loaded pages only), the 'use this first' priority, and a fallback protocol (rephrase, never guess). No contradiction with the read-only, closed-world 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?

Purpose is front-loaded, then scope, then fallback guidance—a logical order. It runs fairly long, but each clause (corpus limit, don't-guess fallback) carries distinct operational value, so little is wasted.

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 an output schema present, return values needn't be explained, and the description correctly focuses on scope, priority, and not-found behavior. For a single-parameter search tool with strong annotations, this is nearly complete; only the missing differentiation from 'search_pages' holds it back.

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% and there is a single 'query' parameter, so the schema already fully documents it (including an example). The description adds no extra syntax, format, or matching semantics for the query string beyond the rephrasing advice, so 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 (探す/検索) plus a precise resource (読み込んだページの中 from a named site) and the return shape (抜粋と URL). However it does not differentiate itself from the sibling 'search_pages', which sounds near-identical, leaving the agent to infer the distinction from the description alone.

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 when-to-use ('質問されたら、まずこれを使ってください') and when-not-to-use constraints ('探せるのは読み込んだページだけ', no internet, no post-load updates), plus recovery guidance (rephrase, then admit the site doesn't cover it). It stops short of naming an alternative sibling to route to, so it lands at 4 rather than 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 declare readOnlyHint=true, destructiveHint=false and openWorldHint=false, so the safety profile is covered. The description adds meaningful non-annotation context: it is a deprecated alias kept for backward compatibility and returns a different shape than search.

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-loads the deprecation/legacy status and the alternative, then adds the return-shape difference. Three compact sentences with essentially no 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?

For a legacy alias with full read-only annotations and no output schema, this is largely complete: it covers identity, relationship to `search`, and migration guidance. The one gap is that it says the return format differs without saying how, which matters when an agent must parse the result.

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 50%: `query` is documented in the schema but `limit` (default 5, max 20) is not. The description mentions neither parameter nor any query syntax or limit semantics, so it fails to compensate for the coverage gap.

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 this is the legacy-named feature that searches within the site, and explicitly distinguishes itself from the sibling `search` by noting the return shape differs. It is specific about verb and resource, though the title still labels it ambiguously as '以前の名前' (previous name).

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?

It gives an explicit routing directive: '新しく使うときは search を使ってください' (use search for new usages) and explains it is retained for existing integrations. That is clear when-to-use guidance, though it does not explain any scenario where this tool remains the correct choice.

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 readOnlyHint=true, destructiveHint=false and openWorldHint=false, so safety is covered structurally. The description adds what comes back (site type, description, page count, main pages), which is genuinely useful since no output schema exists, but it says nothing about freshness, caching, or auth/scope requirements.

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 compact clauses: identity/domain, returned content, and the invocation trigger — front-loaded with the return contents. No filler, though the quoted site title is slightly redundant padding before the domain.

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 zero-parameter read-only tool with no output schema, the description supplies both scope (which site) and expected return fields, which is close to complete. It stops short of explaining how the returned 'main pages' relate to sibling tools like list_pages.

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?

The tool takes zero parameters, so per calibration the baseline is 4. The description appropriately spends no words on parameter semantics, instead naming the single site the tool covers, which is the only input-like information an agent needs.

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 (返します) and resource (site overview for the named domain), then enumerates what it returns: type, description, loaded page count, main pages. It reads as a summary/introspection tool distinct from fetch/get_page/list_pages, though it never names those siblings to sharpen the distinction.

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 trigger: use it when you first want to grasp the overall picture of the site (最初にサイトの全体像をつかみたいとき). That is clear context for invocation, but it names no alternatives or when-not-to-use conditions (e.g. use search_pages for a specific query instead).

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