HORIZON SHIELD — construction estimate auditor (agent intake)

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

Beyond the annotations (readOnly, idempotent, non-destructive), the description discloses deterministic keyword-table routing ('not an LLM'), deterministic outcomes for same input, the presence of verify and limits in every reply, and explicit behavior for unsupported inputs ('does not answer - it returns the appropriate outside destination instead'). This is rich behavioral context that far exceeds the structured annotations, with no contradiction.

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 with headings and examples, and the core English content is front-loaded. However, the Japanese paragraph at the end largely duplicates the English content (scope, routing, out-of-scope behavior, verify/limits), making it redundant and longer than necessary for an agent-facing description. This duplication costs tokens without adding new information, so it is less concise than ideal.

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?

Given the tool's complexity as a deterministic router with four destinations, the description is unusually complete: it covers scope, exclusions, routing determinism, response envelope (verify/limits), and example queries. The presence of an output schema means return-value details need not be repeated. No important gaps are evident for selecting and invoking this 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?

Input schema coverage is 100%, so the baseline is 3. The description adds meaningful usage semantics beyond the schema by explaining the tool is a single entry point that takes 'one sentence,' and provides three explicit examples demonstrating how to phrase the `ask` parameter (e.g., 'is 800,000 yen high for exterior wall painting?'). It also clarifies optional parameters indirectly ('pass one sentence'), which helps the agent construct valid invocations. However, it does not detail each parameter beyond what the schema already documents.

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 opens with a specific verb+resource statement: 'Check whether a construction or renovation quote in JAPAN is fairly priced...' and enumerates additional use cases (verify HORIZON SHIELD receipt, jidec citation, overcharging tactics, verified contractor). It clearly distinguishes scope from siblings via explicit DO NOT USE conditions, and examples solidify the intended purpose.

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 provides explicit USE WHEN and DO NOT USE WHEN sections, covering exclusions such as outside Japan, non-construction, post-signing, and emergencies. It lacks direct named alternatives, but states that out-of-scope inputs 'return the appropriate outside destination instead,' which implicitly routes to other tools. This is clear context but stops short of naming sibling alternatives, so it merits a 4 rather than 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?

The annotations already indicate read-only, idempotent, and non-destructive hints. The description adds crucial behavioral details: it only generates drafts, does not auto-post, includes backlinks, and restricts to verifiable first-party data. This goes beyond the annotation hints, providing transparency about the tool's side effects.

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 concise, with the main verb and object front-loaded. It includes both Japanese and English versions, but all sentences carry useful information about constraints and workflow.

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?

Given a simple one-parameter tool with an output schema and rich annotations, the description sufficiently covers the tool's purpose, constraints, and workflow. It explains the draft nature and operator finalization, making it complete for an agent to select and invoke the 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?

The input schema has one parameter 'work' with a complete description including examples. The tool description does not elaborate on the parameter, but with 100% schema coverage, the baseline of 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 clearly states the tool generates broadcast drafts (note long-form and X short-form) with backlinks to HORIZON SHIELD explanation pages, using only verifiable first-party prices. This specific verb+resource combination distinguishes it from sibling tools like ask, intake_estimate, orchestrate, and scan_tactics.

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 implies usage context: when a draft alert about construction overcharging is needed. It clarifies constraints (only KIRA data, no guessed numbers) and states that the operator finalizes, but doesn't explicitly mention alternatives or when not to use it. Since there are no exclusions, a score of 4 is appropriate.

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=true, idempotent=true, and destructive=false. The description adds valuable nuance by stating that it does not make definitive price judgments ('価格の断定はせず') and instead hands over points to check, plus a path to a third-party check. This goes beyond the annotations and does not contradict them.

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 economical and front-loaded, but it duplicates the content in Japanese and English, which doubles length. The English translation adds accessibility but is redundant; still, there is no filler and the essential information is present.

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?

Given the simple two-parameter input, a rich output schema, and full safety annotations, the description covers the tool's purpose, non-assertive behavior, and return pathway. It does not discuss error cases or exclusions, but these are not essential given the structured metadata already present.

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?

Both parameters (work, quoted_price) are fully described in the schema (100% coverage). The description only restates the parameter concepts ('工事名と業者提示額') without adding extra syntax, format, or behavioral details, so it meets the baseline for schema-heavy documentation.

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 identifies the tool as an intake desk for homeowner construction/renovation quotes, bridging to the KIRA fair-price audit and returning a verdict plus a free third-party check path. It uses specific verbs ('受け付け', '橋渡し') and distinguishes its role from the sibling tools by focusing on quote intake and diagnosis.

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 provides clear context: use this tool when a homeowner has a quote that needs a fair-price diagnosis, and it returns a verdict plus a free EHN path. However, it does not explicitly say when not to use it or mention alternatives among the sibling tools, so it falls 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 declare readOnly, openWorld, idempotent, and non-destructive hints. The description adds valuable behavioral detail: it internally executes subtools in order, restricts prices to verifiable first-party data, and never auto-posts (drafts only). 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 bilingual but efficient; each sentence carries meaning: the orchestration role, internal flow, pricing constraint, and no-auto-posting. The English summary is slightly redundant but not wasteful. Front-loaded with the core purpose.

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?

Despite having an output schema (which covers return format details), the description still tells the agent what the result contains (audit + tactics + broadcast draft). It explains the orchestration sequence, subtools, and key constraints, making it highly complete for a multi-step 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 coverage is 100%, with both work and quoted_price fully described (including the effect of quoted_price triggering KIRA diagnosis). The description restates this but adds no new parameter semantics beyond the schema, so baseline 3 is appropriate.

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 identifies the tool as a one-call orchestrator that sequentially runs intake, scan_tactics, and draft_broadcast, then bundles the results. The verb 'orchestrate' plus the explicit pipeline distinction makes it stand out from sibling tools, which are individual steps.

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 states it runs the full pipeline in a single call, implying use when the complete flow is needed. It also clarifies constraints (verifiable prices only, draft-only output). However, it does not explicitly tell the agent when to call individual sibling tools instead of this orchestrator.

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?

The description adds meaningful behavioral context beyond the annotations: it states the data is verified (from HORIZON SHIELD with 30 years of supervision), that it points to primary sources, and importantly, it will not assert new cases via speculation ('推測で新事例を断定しない'). These caveats disclose limitations and behavior not captured in the readOnly/openWorld/idempotent hints, enhancing transparency.

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 exceptionally concise: two sentences (plus a brief English summary) that front-load the main action ('Returns verified overcharge tactics...') and follow with key caveats. Every word earns its place, with no repetition of schema information or annotations.

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 tool with a rich annotation set and an output schema, the description is highly complete. It covers purpose, data source, what it returns (tactics and source locations), what it is not (price verdict), and its epistemic boundary (no speculation). No critical aspect is missing for effective use.

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 input schema already provides a full description of the 'work' parameter (100% coverage) with examples in Japanese. The description merely echoes 'ある工事・キーワード' without adding any new semantics or constraints beyond the schema. Therefore, it meets the baseline but does not exceed it.

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's verb and resource: 'Returns verified overcharge tactics and points to primary sources.' It specifies the input (a construction/keyword) and the source (HORIZON SHIELD's verified data), and distinguishes itself from price judgment by noting 'Awareness, not a price verdict.' This makes it unmistakable and distinct from sibling tools like ask or draft_broadcast.

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 implies when to use the tool: for a given construction/keyword when you need verified overcharge tactics and primary sources. It also specifies an exclusion: 'not a price verdict,' indicating that it should not be used for price evaluation. However, it does not explicitly name alternative tools, so it lacks the highest level of guidance.

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