Kitbench

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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive), so the bar is lower. The description adds genuine domain behavior the annotations cannot: exactly which days 'business' skips, the N range of 0–100,000, and that output includes weekday and wareki. No auth or rate-limit context, so not a 5.

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 verb+resource before the unit nuances. Slight redundancy against the schema (the 0–100,000 range and unit semantics are repeated), but nothing 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 5 parameters, no output schema, and full schema coverage, the description gives enough to invoke correctly: direction, units including holiday handling, and lang. It does not detail the full return shape, but the key returned fields (weekday, wareki) are flagged, which is sufficient.

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 every parameter is already documented in the schema, and the description's unit and N-range text largely restates it. Baseline 3 is appropriate; the description adds only marginal meaning beyond the structured fields.

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 ('Find the date N days after or before a date') and even names the returned fields (weekday and Japanese era date). That cleanly separates it from siblings like days_between (a difference, not an offset) and convert_japanese_era (era conversion only), so an agent can route 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 real conditional guidance: 'business' for Japanese holidays, 'weekdays' for weekends only, plus the 'Good for deadlines in Japan' framing that tells the agent when this tool is the right pick. It stops short of explicitly naming an alternative tool to use instead, so 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 readOnly, idempotent, non-destructive, non-open-world, so the safety profile is covered. The description adds genuinely new behavioral context beyond the schema: the Feb 29 -> Mar 1 rule and the Japan-timezone default for the reference date, which an agent could not infer otherwise.

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 verb and purpose, followed by a dense output list and two short edge-case sentences. Every clause carries information, though the long mid-sentence output enumeration is list-heavy rather than prose-optimized.

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 enumerates the returned fields and covers the key edge case (leap-day) and the reference-date default. The only thin spot is that lang's effect is left entirely to the schema, but overall it is complete enough for correct invocation.

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 documents all three parameters including the lang enum and date range. The description adds only one semantic not in the schema, the default reference date being today in Japan, so 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 opens with a specific verb+resource ('Calculate exact age from a birth date') and then enumerates the concrete outputs, so an agent knows precisely what the tool computes. It is easily distinguished from siblings like convert_japanese_era or days_between because it bundles multiple derived values rather than performing one conversion.

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 is only implied: the leap-day rule and the 'reference date defaults to today in Japan' note signal relevant contexts, but there is no explicit statement of when to pick this over days_between or convert_japanese_era. The guidance is adequate but leaves the routing decision 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?

Annotations already cover the safety profile (readOnly, idempotent, non-destructive), and the description adds genuinely useful context beyond them: exact integer arithmetic, the returned fields (tax amount, both prices, formula), and the 13-digit yen ceiling. It does not discuss error handling for invalid rates or amounts, which keeps it from a 5.

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?

A single dense sentence that front-loads the core operation and then the parameter options, with no filler. It is slightly overloaded with enumerations, but every clause carries 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 usefully names what is returned, and it documents all parameters plus the numeric limits. An agent has enough to call it correctly; only error/failure behavior 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 description coverage is 100%, so all four parameters are already documented in the schema; the description largely restates the rate and rounding choices. It adds little syntax or format detail the schema lacks, so 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?

States a specific verb and resource ('Calculate Japanese consumption tax'), spells out both conversion directions, and makes clear this is a tax calculator rather than any of the date/unit/conversion siblings. An agent can identify it immediately without opening the 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?

Explains the two operating modes (tax-excluded to tax-included and the reverse) and enumerates the rate and rounding options, which effectively tells the agent when each parameter path applies. It stops short of explicit exclusions or 'use X instead' routing, but the sibling tools are unrelated utilities, so no routing is really needed.

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 cover the safety profile (readOnly, idempotent, non-destructive), so the description's added value is the validation behavior: it rejects dates that do not exist in an era (Heisei 31-05-01) and enforces a 1873-01-01 to 2200-12-31 range. Those are concrete failure conditions an agent needs, beyond what annotations state. Return format is still undisclosed.

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 carrying a distinct payload: the conversion contract, the accepted input formats, then validation/range limits. Purpose is front-loaded and there is no filler, though the format-example sentence is dense enough that a small amount of scanning is 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?

For a converter with no output schema, the description explains inputs thoroughly but never says what comes back (both representations? era name? gannen rendering?). The lack of required params also means the agent must infer that exactly one input mode should be supplied, which is only implied.

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?

With 67% schema coverage, the schema documents era, date, year and wareki, but the description adds the accepted wareki syntax variants and the 'gannen' (元年) semantics for year 1. It also clarifies that era/year/month/day is a combined alternative to the single date or wareki inputs, which the schema does not express.

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 bidirectional verb+resource: 'Convert between Western (Gregorian) dates and Japanese era dates'. The parenthetical naming of the five eras (Meiji, Taisho, Showa, Heisei, Reiwa) further pins down the domain, and no sibling (add_days, calculate_age, days_between) overlaps with this conversion task.

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 explicitly tells the agent the three acceptable invocation shapes: a YYYY-MM-DD date, a wareki string like '令和元年5月1日' / 'R1.5.1', or era + year + month + day. That is real routing guidance for a tool with six optional parameters and no required ones. It does not name exclusions or alternatives, which keeps it 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/idempotent/non-destructive, so the safety profile is covered. The description adds real context beyond that: exact factors from official sources, and a concrete picture of the return payload (exact result, rounded display value, factor, plus extras like feet-and-inches, 1/16 inch fractions, pounds-and-ounces). It does not mention error behavior for mismatched categories.

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, factors and return shape are front-loaded in the first two sentences. The trailing unit-id block is long and largely duplicates the schema enums, but its category grouping earns its place by encoding valid pairing constraints; only mild redundancy remains.

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-value burden and does so concretely (exact value, formatted value, factor, extras). Inputs are fully enumerated and grouped. Missing only edge-case behavior such as cross-category inputs or out-of-range significant_digits handling.

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% (baseline 3), and the description goes beyond it by grouping the flat enum of unit ids into named categories (length, mass, temp, volume, area, speed, fuel, pressure, energy, power, data), which is the information an agent needs to pick a valid from/to pair. This is value the schema does not convey on its own.

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 (convert units) and immediately scopes it with distinguishing detail: official exact factors (NIST), Japanese traditional units, temperature and fuel economy. No sibling tool does unit conversion, so the agent can select this unambiguously.

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 is implied by the domain rather than stated; there are no explicit when-to-use/when-not statements or references to alternatives (none exist among the date/tax/text siblings). The description does imply the from/to must share a category by grouping unit ids by category, which is helpful implied guidance, but it never says what happens on a cross-category pair.

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), and the description adds genuinely new behavioral detail: how grapheme clusters, combining marks, flags and half-width characters are treated, and the 200,000-character ceiling. It does not describe performance or error behavior, keeping it short of a 5.

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 counting semantics are front-loaded in the first sentence, with the enumerated output list and the size ceiling following. The single dense paragraph is all signal, though the long metric list is slightly unwieldy.

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?

There is no output schema, so the description must carry the return-value burden, and it does: it enumerates grapheme counts, newline/whitespace-excluded counts, half-width weighting, lines, paragraphs, genkō yōshi sheets, UTF-8 bytes, English words and script breakdown. Nothing an agent needs to call or interpret this tool 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 100%, so both parameters are already documented by the schema, including the limit parameter's remain/over behavior. The description reinforces the input size constraint but adds no syntax or format meaning beyond what the schema states, making the baseline 3 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?

States a specific verb (count) and resource (characters in a text) and enumerates the exact metrics returned, so the agent knows precisely what this tool produces. The counting semantics (grapheme clusters, half-width as 0.5) clearly separate it from the unrelated date/unit/tax 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?

The phrase 'Strong for Japanese text and character limits' implies the appropriate use case, and sizes are bounded ('up to 200,000 characters'). However, there is no explicit when-to-use vs when-not guidance and no alternatives are named, leaving usage 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?

Annotations already declare this as a read-only, idempotent, non-destructive, closed-world operation. The description adds substantive behavioral context on top: exact calendar arithmetic with no time-zone handling, the valid date envelope (1873-01-01 to 2200-12-31), and the dependency on the Cabinet Office holiday list for Japanese business days.

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 tight sentences, front-loaded with the core verb and the list of produced measurements, followed by precision caveats (no time zones) and the valid range. No redundant or filler phrasing.

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 what is returned (total days, weeks/days, years/months/days, Japanese business days, weekdays) and carries the time-zone and date-range constraints. It does not clarify the return shape (single object vs. multiple fields) or the inclusive/exclusive default of the end date, leaving a small 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 description coverage is 100%, so all four parameters (including the lang enum and include_start_day semantics) are already documented. The description only restates the optional start-day inclusion and the date range, adding no format or edge-case meaning beyond the schema. 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 ('Count the days between two dates') plus the exact measurement modes it produces, which clearly separates it from siblings like add_days, calculate_age, and convert_japanese_era. An agent can identify the resource and scope without opening the 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?

Usage is implied by the enumeration of measurement modes (business days, weekdays, weeks/days), but there is no explicit statement of when to choose this over add_days or calculate_age, nor any exclusion guidance. A competent agent can infer the fit, but the routing signal is weak.

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