b612 (hakuto AI)

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

With no annotations, the description carries the full burden and does well: it enumerates the returned content (禁止事項, 必須事項, 6つの不足パターン, 貼れるCSS/JS一覧) and the mode switch where supplying a snippet returns the full implementation. It does not state the operation is read-only or discuss freshness/staleness of the rules, but for a lookup tool the disclosure is substantial.

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 imperative trigger is front-loaded, followed by the returned categories and then the parameter behavior. It is dense and example-heavy (the parenthetical list of AI-like shapes), but each clause maps to real returned content rather than 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 one optional parameter, no annotations, and no output schema, the description must describe returns itself — and it does, listing the rule categories, the six missed patterns, and the snippet mode. An agent can call this correctly without further information; only the sibling-routing question remains open.

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 parameter is already documented with format (file name examples) and the omit-behavior. The description's "snippet に名前を渡すとその実装を全文返す" restates the schema rather than adding new meaning, 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 states a concrete verb (引く/返す) and a specific resource: design conventions for HP/LP/UI/チラシ, covering prohibitions, must-dos, AI-missed patterns, and CSS/JS snippets. It is clear what an agent gets back. It does not, however, differentiate itself from siblings like b612_rules, b612_principles, or b612_patterns, whose names suggest overlapping rule/principle content.

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?

"ホームページ・LP・UI・チラシを作る前に必ず引く" gives an explicit trigger condition, and the final sentence states when to pass a snippet name. No when-not conditions or named alternatives (e.g., b612_review_checklist) are given, so the routing between siblings is left to inference.

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?

With no annotations, the description carries the full behavioral burden. It discloses the nature of the output (a reference bundle of models, edge cases, laws, anti-patterns) and implies a read-only lookup, but says nothing about freshness, auth needs, rate limits, or whether unknown domains error or fall back.

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 purpose clause is nicely front-loaded in one sentence, but the entire second sentence duplicates the domain enumeration already present in the input schema, adding bulk without new information.

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 must stand in for return semantics; it names the four content categories it returns, which is enough to set expectations, but gives no indication of the response shape or size for the 36 supported domains.

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 required parameter is already documented in the schema, so the baseline is 3. The description merely restates the same domain list and adds no format, casing, or defaulting detail beyond what the schema provides.

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 gives a specific verb (返す / returns) plus a concrete inventory of what is returned (must-have models, edge cases, required laws/regulations, anti-patterns) scoped to a domain. It is clear what the tool produces, but it never distinguishes itself from close siblings such as b612_patterns or b612_principles, which plausibly return overlapping knowledge content.

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?

There is no when-to-use statement, no prerequisite, and no routing guidance relative to the seven sibling tools. The domain list is a parameter enumeration, not usage guidance, so the agent must infer the trigger condition on its own.

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?

With no annotations, the description carries the full burden and does well: it discloses the default window (1 day), and explicitly scopes downstream work away from the tool — summarization, classification and task-ification must be done by the caller on the raw result. It does not mention auth/permission requirements for reading the LINE account or any rate limits, leaving a small gap.

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?

Leads with what is read, then trigger phrases, then the parameter, then the post-processing constraint. No filler; each sentence carries actionable information.

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?

Even without an output schema, the description states what comes back (raw memos ordered by date), which is the key unknown for the caller, and it bounds the parameter. Nothing an agent needs to call this correctly 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 description coverage is already 100%, and the schema documents the default of 1 itself. The description restates 'days で遡る日数を指定する（既定は1＝今日ぶん）' and adds the mapping '今週なら 7', which is marginal added value over the schema — baseline 3 applies 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?

Specific verb (読み込む) plus a precisely scoped resource: memos the user sent to hakuto's official LINE. This is clearly distinguishable from every sibling (design, rules, patterns, review_checklist, update), none of which read user-submitted notes.

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 trigger utterances (「今日のメモをまとめて」「昨日のメモ見せて」「今週のアイデア整理して」) that map directly to invocation, which is strong when-to-use guidance. It does not name an alternative tool or state when NOT to use it, 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?

No annotations exist, so the description carries the full burden. It does disclose useful non-obvious traits: this is a 公開版 with no 案件固有名, the full text is ~88,000 characters, and truncation names the dropped sections. It does not state that the tool is strictly read-only, nor anything about permissions or rate limits — modest but real 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?

Four sentences, front-loaded with what is returned and followed by output-shape and parameter behavior. Every sentence carries information; there is no filler. Minor overlap with the schema's parameter descriptions keeps it from a 5.

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 parameterless-required, read-only retrieval tool with a fully documented schema and no output schema, the description covers what an agent needs: default behavior, size constraints, truncation semantics, and the recommended query path. The only gap is that it never describes the format of the returned index or matched sections.

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 baseline is 3. The description goes beyond the schema by explaining the motivation behind the parameters — the corpus is too large to return whole, which is why full defaults to false and query is recommended — and by giving concrete query examples of the domain vocabulary. That rationale adds genuine meaning over the field-level text.

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 gives a specific verb and resource — it returns b612 の実戦ノウハウ and enumerates the content types (典型エラー・推測実装の傾向・過去に効いた意思決定). That is far more concrete than the bare name 'patterns'. It does not, however, distinguish itself from close siblings like b612_principles, b612_rules, or b612_notes, leaving the agent to infer which knowledge base to consult.

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 clearly explains the tool's own operating modes: default returns only the heading index, and passing a query returns just matching sections, explicitly marked 推奨. What it never addresses is tool selection — no statement of when to prefer this over b612_principles, b612_rules, or b612_notes. Usage is implied rather than routed.

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?

With no annotations and zero parameters, the description carries the full behavioral burden. The verb '返す' implies a side-effect-free read and it discloses the contents and the sequencing intent, but it never explicitly states that this is a read-only, no-side-effect call and does not describe the form of the returned content (single text blob vs. structured list).

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, both earning their place: the first enumerates the payload, the second states the timing and follow-through. Content is front-loaded with the highest-value information.

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, no-output-schema retrieval tool the description is nearly sufficient: it names the content and the when-to-call. The only residual gap is sibling disambiguation against b612_rules / b612_patterns / b612_review_checklist, which an agent might otherwise confuse.

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 no parameters, so there is no parameter semantics to document; the baseline of 4 applies. The description correctly adds no misleading param details.

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 (返す/returns) and an enumerated resource (NO SPECULATION, 実装前の3問自問, 5原則, 握りつぶし禁止), so an agent knows exactly what content arrives. It does not, however, distinguish this from near-siblings like b612_rules or b612_patterns, so the boundary is left to inference.

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 explicit timing guidance: read this before starting a review or implementation and proceed along the principles. That is clear context for invocation. It stops short of naming when NOT to use it or pointing to an alternative sibling.

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?

No annotations, so the description carries the full burden. It does disclose the division of labor (this returns data; the caller does the reviewing), which usefully implies a read-only lookup, but it says nothing about error behavior for an unsupported type or the shape/size of the returned checklist.

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 short sentences, front-loaded with the core action and ending with the supported values. No filler, though the trailing value list duplicates the schema enum.

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 lookup with no output schema, the description covers what is returned, who consumes it, and the accepted inputs. Only the omission of the return structure and unsupported-type handling keeps it from being fully 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% and the schema already lists the accepted target_type values with a fallback to generic. The description repeats that same list rather than adding syntax, format, or fallback semantics beyond what the schema provides, 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+resource: returns b612 review perspectives keyed by target type, and explicitly says the caller's AI uses them to run the actual review. It doesn't differentiate itself from similar-sounding siblings like b612_principles or b612_rules, which an agent could confuse with a checklist feed.

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 gives implied usage context — pass a target type, get perspectives, then perform the review yourself. But it names no alternatives and gives no when-not guidance against the other b612_* retrieval tools.

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?

With no annotations, the description must carry the behavioral burden, and it does disclose the key trait that no code is transmitted to the server — an important privacy/side-effect signal. It also specifies the returned fields. It omits auth, caching, or versioning behavior, but for a zero-parameter static catalog those are largely irrelevant.

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 purpose is front-loaded in the first sentence, but the description then appends a raw list of roughly 32 rule IDs. Since the tool itself returns those IDs, the enumeration is largely redundant token weight, even if it does signal coverage breadth. The category summary (swallowed errors, hardcoded secrets, destructive SQL, XSS, etc.) already conveys scope more economically.

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, parameterless catalog tool with no output schema, the description covers what is returned, the fields of each rule, and how the caller is expected to use it. An agent has enough to select and invoke it correctly; only the sibling-differentiation gap keeps it from full completeness.

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 the rubric the baseline is 4. There is no parameter surface for the description to explain or for the schema to under-document.

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: it returns a catalog of detection rules with their ID, severity, rationale, and fix. The added explanation that the calling AI applies the rules locally makes the operating model concrete. It does not name any sibling (e.g. b612_review_checklist or b612_principles), so the agent must still infer the boundary itself.

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 gives clear context for how the tool is meant to be used: fetch the ruleset, then apply it to your own local code, with code never being uploaded to the server. That workflow guidance is genuinely useful. It provides no explicit when-not-to-use conditions or named alternatives among the b612 siblings.

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?

With no annotations, the description carries the full burden and does so well: it discloses that the output is executable commands, that only differing files are replaced, and that existing files are backed up as .bak. It omits permission/auth requirements and any no-op/error behavior, but the mutation semantics are unusually well surfaced for a description-only 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?

Front-loaded with the purpose, then use-cases, then effect of execution. Sentences each carry distinct information, though the enumeration and repeated trigger phrasing make it slightly longer than strictly needed.

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 and no annotations, so the description must explain the return value and side effects — which it does (returns commands; executing them swaps only differing files with .bak backups). What remains unstated, such as behavior when nothing needs updating, is minor for a 1-param procedural tool.

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 sole parameter (base_url) is optional and already documented ('接続先（通常は指定不要）'). The description adds no parameter-level detail, 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 (b612 のアップデート) and enumerates the asset classes touched (エージェント・法令ファイル・業種プリセット・雛形・参照ドキュメント). Critically it clarifies the tool returns a procedure/commands rather than performing the update itself. Sibling differentiation is only implicit, but the scope is unambiguous.

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?

Provides explicit trigger phrases (「b612 をアップデートして」「b612 を最新にして」) and two concrete triggering situations (新エージェント追加時、法令改正時). It does not name a when-not condition or a routing alternative among the sibling b612_* tools, 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.