台灣健康資訊

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

With no annotations, the description carries the full behavioral burden. It usefully discloses that the result is a target-range judgement plus hypoglycemia handling, and clarifies the mg/dL unit, but says nothing about out-of-range response behavior, validity bounds for the value, or whether any action is recommended beyond treatment guidance.

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 action and resource before the parameter and return details. No filler, though the two clauses could be split for faster scanning.

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 two-parameter interpretation tool with no output schema, the description covers purpose, both parameters, and the nature of the returned values. Only the numeric validity range and threshold specifics are left implicit, which is minor for this tool class.

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 compensate, and it does: it gives the unit for value (mg/dL) and supplies the semantic meaning of both enum members (fasting=飯前, after_meal=飯後), which the bare enum values do not convey. It does not specify an acceptable numeric range for value.

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 (判讀/interpret) and resource (單次血糖值, single glucose reading), names the unit (mg/dL), and describes the return (whether within general control target plus hypoglycemia handling). No sibling shares this resource domain, so an agent can distinguish it immediately from check_home_blood_pressure and the cancer-related 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?

No when-to-use, when-not-to-use, or prerequisite guidance is given. The resource domain differs from every sibling, which provides implicit routing, but the description never states the scenario in which interpreting a single reading is appropriate versus the other check_* 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?

No annotations exist, so the description carries the full disclosure burden. It conveys the input-driven, eligibility-listing nature and the year-115 program scope, but omits result format, any age/eligibility boundary behavior, and whether partial inputs yield partial results. Adequate but thin for a no-annotation 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?

A single compact sentence with the resource and screening list front-loaded and no filler. It is efficient, though the parameter enumeration is embedded rather than structured.

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 an 8-parameter tool with no annotations and no output schema, the description should clarify the less obvious inputs (pack_years, indigenous) and at least sketch the return shape. It covers the broad strokes but leaves meaningful gaps around the most decisive lung-screening parameters.

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 descriptions must compensate. The description names age, sex, smoking, betel and 'family history' (loosely covering family_lung/family_colorectal), but never explains indigenous status or pack_years, both of which materially affect lung-screening eligibility. Partial compensation only.

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 (list) and resource (Taiwan publicly funded cancer screenings) with the six screening categories enumerated, so the agent knows exactly what comes back. It does not name or contrast any sibling, but the siblings (blood glucose, blood pressure, meal risk) are unrelated enough that differentiation is not strictly needed.

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 phrasing 'based on age, sex, smoking, betel, family history' tells the agent these inputs drive the lookup, but there is no explicit statement of when to prefer this tool or when it is inappropriate (e.g., for non-Taiwan residents, or ages outside the program).

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 disclosure burden, and it does list the computed outputs (average, grading, critical-value determination, 722 method), which is meaningful behavioral context for a statistical/logic tool. It stops short of noting edge cases such as malformed input or how critical values are thresholded.

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 densely packed sentence front-loads the action and guideline, then appends the output list and input example. No filler; every clause carries operational 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 one-parameter computation with no output schema or annotations, the description covers purpose, standard, output contents, input format, and input limit. It is nearly sufficient, missing only edge-case/error behavior that an agent might need.

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 0% and there is no description on the single 'readings' property, so the description must compensate. It supplies the exact input syntax (128/82,134/86) and the cardinality limit (up to 6 entries), which is precisely the information the empty schema omits.

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 (interpret home blood pressure) plus the governing standard (Taiwan 2022 guideline 130/80), and enumerates what it produces: average, classification, critical-value flag, and 722 measurement method. This clearly distinguishes it from siblings like check_blood_glucose, which operate on a different metric.

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 its context (interpreting at-home readings against a specific guideline) but never states when to choose this tool over an alternative or what preconditions apply. No exclusions or sibling routing are given, so usage is only inferable from the name and content.

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. It does disclose the processing logic (IARC comparison, prevention guidelines) and the shape of the output (risk items plus improvement suggestions), which is more than nothing. However, it says nothing about whether the input is stored, whether it is purely informational vs. medical advice, or any limitations of the assessment.

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 sentence structured as input format first, then processing, then output contents. Front-loaded and free of filler, though it is dense enough that it could be split for readability.

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 one-parameter, annotation-free, output-schema-free tool, the description covers input format, processing basis, and output categories. The main residual gap is behavioral framing (advice vs. diagnosis, data handling), which is minor for this tool's scope.

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 0% and there is a single 'meal' parameter, so the description must compensate. It does, by specifying the expected content is natural-language food description in Chinese and that the granularity can be one meal or a whole day, which the bare string schema does not convey.

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 names a specific action (compare a meal against IARC carcinogen tiers and prevention guidelines) and a concrete output (risk items like processed meat, red meat, alcohol, betel nut plus improvement advice). It is clearly distinct from siblings such as check_blood_glucose or check_cancer_screening_eligibility, though it never names a sibling explicitly.

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 by the input phrasing ('input what you ate for a meal or a day'). There is no statement of when to prefer this over get_counseling_guide or check_cancer_screening_eligibility, and no exclusions or prerequisites.

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 are provided, so the description carries the full disclosure burden, yet it says nothing about the tool's behavior: no mention that it returns static reference text with no side effects, no auth requirements, and no return format. The topic list describes content, not 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?

A single dense sentence with no filler, front-loading the core topic (legal conditions for online counseling) before listing secondary items. It is efficient, though the absence of a lead verb makes it read more like a table of contents than a description.

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 info-lookup tool with no output schema, the description's enumeration of the guide's sections (legality, age limits, free program, booking steps, crisis hotline) gives an agent enough to decide whether a call will satisfy the user's need.

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 are no parameter semantics the description needs to compensate for.

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 is a dense noun phrase rather than a verb+resource sentence, but combined with the name get_counseling_guide it clearly enumerates the returned content: Taiwan online counseling legality, the 18-year-old restriction, the free 3-session program for ages 15–45, booking steps, and crisis lines like 1925. An agent can identify the resource, though the description doesn't explicitly differentiate it from sibling health-check tools.

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 explicit when-to-use statement or named alternative, but for a zero-parameter informational lookup the trigger context is implied by the enumerated topics (questions about counseling legality, eligibility and crisis help). Usage is inferable rather than stated.

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 are provided, so the description carries the full burden. It usefully discloses the geographic/regulatory scope (Taiwan-approved drugs), the safety-oriented nature of the content, and an explicit out-of-scope boundary (no dosage advice). It stops short of stating that the tool is read-only and takes no parameters or what the output shape is.

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 tightly packed sentences with the resource front-loaded, followed by a short scope-limiting clause. Every element (drug names, content categories, disclaimer) earns its place; 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 parameterless knowledge-retrieval tool with no output schema and no annotations, the description covers the essential ground: what topic, which drugs, and what categories of information are returned. The absence of any return-format note is a minor gap given this tool type.

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 and the schema is an empty object, so there is nothing for the description to clarify; baseline 4 applies. The mention of specific drug names implicitly maps the query domain but is not parameter-level guidance.

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 precise resource (side-effect profile of Taiwan-approved GLP-1 weight-loss drugs) and enumerates the exact content returned: incidence rates, serious warning signs, when to seek care, and TFDA reminders. Naming the three specific drugs (tirzepatide, semaglutide, liraglutide) makes the scope unmistakable and clearly separates it from the screening/counseling 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?

Usage is implied by the content scope (a user asking about GLP-1 side effects), and the closing exclusion '不提供劑量建議' draws one boundary. However, it never names an alternative sibling or states an explicit when/when-not condition, so routing still relies on inference.

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