PediBot

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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive, closed-world), so the description adds genuinely new behavior: the exact stylistic contract, that medical content comes only from published guidelines, and that triage runs first (a safety ordering guarantee). This is real context beyond the structured fields.

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 in the first sentence, usage routing sits in the second paragraph, and every clause carries information (audience, ages, format, sourcing, audience segments). Slightly dense but 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?

With an output schema present, return values need no explanation, and the description covers audience, style contract, sourcing constraints, triage ordering, and alternative-tool routing. 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 100%, so lang, country, and question are already documented with enum, pattern, and length constraints. The description adds the useful 'eight languages' fact matching the lang enum, but nothing else about parameter syntax or behavior. Baseline 3 is appropriate 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?

The description names a specific transformation (a sourced answer rewritten to be read aloud to a child aged 5-10) with concrete output characteristics: three to five warm sentences, no frightening words, one comparison, one action. It is instantly distinguishable from paediatric_question_with_sources, which it explicitly names.

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 when: 'Use when the answer will be read by or to a child aged 5 to 10.' It then names the alternative and its condition: 'For the parent's own question use paediatric_question_with_sources.' No inference is required.

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 read-only, idempotent, non-destructive and closed-world behavior. The description adds valuable operational context beyond that: severe acute malnutrition is flagged as urgent, and supplying a country makes the tool report which chart that country's health record uses. It does not discuss response format or error conditions, but the output schema exists.

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 compact paragraphs: the first front-loads what the tool computes and its authoritative sources, the second immediately states the usage rule and the alternative. Every clause carries information, with no filler or redundant restatement of the name.

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 six-parameter tool with full schema coverage, an output schema, and rich annotations, the description supplies everything an agent needs: purpose, when to use it, alternative routing, country-chart behavior, and the urgent malnutrition flag. Return values are left to the output schema, which is appropriate.

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 schema already documents each parameter thoroughly. The description still adds meaning by explaining the country parameter's chart-selection behavior and reinforcing that sex, age and at least one measurement are needed. It does not independently document all parameter constraints, but it goes beyond the schema baseline.

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: calculating where a child sits on growth charts from named WHO/CDC tables, covering weight, height, weight-for-height and BMI percentiles and z-scores. It distinguishes itself from siblings by naming paediatric_question_with_sources and contrasting arm-measurement malnutrition queries.

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 says when to use it: 'Use when you have a child's sex, age and weight or height and want the percentile.' It also names the alternative for malnutrition with an arm measurement and general growth questions, leaving no ambiguity about tool selection.

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 read-only, idempotent, non-destructive, closed-world behavior. The description adds important provenance and reliability context: 8 tables hand-transcribed from ministry documents and 67 read from the WHO database, plus source URL and review date. 'A transcribed table, not a recollection of one' further signals evidence-based output.

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 front-loaded with the core purpose and return contents. It is somewhat dense with provenance details, but those details are relevant and the final routing sentence is concise.

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 existing output schema and full schema description coverage, the description supplies all additional context an agent needs: country coverage, data provenance, age-based conditional output, and alternative tool routing.

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 schema already documents lang, country, and age_months with enums and ranges. The description reiterates that age_months triggers 'due now' and 'next' output, but adds no syntax or format detail beyond the schema.

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: the official childhood vaccination schedule for 75 countries. It specifies the return contents (ages, vaccines, protection, issuing body, source URL, review date) and distinguishes itself from paediatric_question_with_sources.

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 guidance: 'Use when asked which vaccines a child gets, or are due, in a given country.' It also names the alternative for a different need: 'For what a vaccine does or its side effects use paediatric_question_with_sources.'

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, closed-world, non-destructive, so safety is covered; the description adds substantive behavior beyond that: numbers come from fixed tables with the model never producing a number, and the tool refuses and refers when the child is too young for home medication. It also discloses the return contents (mg range, ml per strength, interval, daily max, warnings), which the agent would otherwise not know cheaply.

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 dense sentences that are front-loaded with the core action and then the routing guidance. The parenthetical brand list and country example are long but each earns its place by preventing wrong invocations; nothing is redundant padding.

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 the description need not enumerate return values, yet it still summarizes them usefully; combined with the refusal path, the country/brand coverage, and the explicit sibling routing, an agent has everything needed to call this correctly.

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 the baseline is 3, but the description adds real meaning: the breadth of accepted brand names (35 brands / 57 countries) and the reason country matters (it determines which strength is listed, with a concrete Haiti/Dominican Republic example). Weight is tied to the core computation and age is tied to the refusal behavior, linking parameters to outcomes.

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 — computes paracetamol/ibuprofen dose for a child by weight from fixed lookup tables — and explicitly distinguishes itself from siblings by naming paediatric_question_with_sources and oral_rehydration_plan. The scope constraint (only two medicines covered) is also stated.

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 states when to use it ('how much paracetamol or ibuprofen to give a child') and routes the agent elsewhere for two distinct cases: fluids in vomiting/diarrhoea to oral_rehydration_plan, any other medicine to paediatric_question_with_sources. Names the required input (weight).

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 it is a read-only, idempotent, non-destructive, non-open-world tool, so the safety profile is covered. The description adds that these are fluids not medicine, that figures come from leaflets never from a model, and that it uses named sources (AEMPS, SEUP, Dioralyte). It does not discuss whether output varies by language or how conflicting sources are resolved, but it goes beyond 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?

Front-loads purpose, then sources, then usage condition and escalation route. Every sentence earns its place; no repetition of schema or annotation content.

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 read-only guidance tool with an output schema and fully documented parameters, the description covers purpose, sources, usage condition, and safety escalation. Nothing an agent needs to call it 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 100%, so the schema already documents each parameter, including enum values and defaults. The description adds that only the age is needed and implies age-band branching, but does not add syntax or format beyond the schema. Baseline 3 is correct when schema does the work.

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: how much oral rehydration solution to offer, volume per intake, frequency, and what to do if the child brings it back up. It distinguishes itself from siblings by naming paediatric_warning_sign_check as a prerequisite in a specific scenario.

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 states when to use it (child vomiting or has diarrhoea), what input is needed (only age), and the exact condition to use a different tool first (signs of dehydration or blood). This is clear when/when-not guidance with a named alternative.

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, closed-world), and the description adds meaningful extra context: empty list on no match, guides written only from published guidelines with named organisations behind each clinical sentence, and eight supported languages. It does not discuss ranking confidence or fallback behaviour when a weak match exists, so it stops short of full 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?

Front-loaded with the core behaviour and well ordered, but slightly padded: the 'hand a parent something to read rather than a paragraph' framing and the repeated mention of eight languages (already an enum) consume words without adding selection signal.

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, the description needn't explain return values, yet it still summarises them harmlessly. It covers usage, routing, languages, and empty-result behaviour, leaving only ranking/confidence semantics unstated, which is a minor gap for a lookup 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%, so both parameters are already well documented in the schema, including the short-noun-phrase guidance and the enum for lang. The description only restates 'in the requested language', adding nothing the schema does not already say. 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 (finds), resource (PediBot's published parent guides), scope (up to five, best match first) and return shape (title, topic, link). It also distinguishes itself from paediatric_question_with_sources by naming the alternative, so an agent can route without opening either 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?

Explicit when-to-use ('when the user wants something to read or share on a topic') and an explicit alternative for the contrasting case ('To answer a specific question use paediatric_question_with_sources'). Both the trigger and the exclusion are 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?

The annotations already declare readOnly, idempotent, non-destructive, non-open-world behavior. The description adds substantial non-obvious behavior: rule-based triage before the model, emergency flagging with the country's number, doses from fixed tables only, and refusal rather than guessing when sources do not support an answer.

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 two purposeful paragraphs: the first front-loads scope, sources, and safety guarantees; the second provides routing rules. Every sentence earns its place and an agent can decide whether to call it without reading further.

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 annotations covering safety traits and an output schema presumably handling return values, the description is complete for correct selection and invocation. It covers source policy, language scope, emergency handling, dose handling, refusal behavior, and sibling alternatives.

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 schema already documents question, lang, and country with examples, enum, and purpose. The description restates language support and the country's role in emergency numbers but adds no new syntax, format, or validation detail beyond the schema, so the 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 states a specific verb+resource (answers a parent's paediatric question from published guidelines) and immediately names the supported source families and languages. It distinguishes itself from siblings by explicitly routing dose, vaccination, and danger-screening questions elsewhere.

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 trigger: use when a parent asks a free-form question about a baby's or child's health. It then names three sibling alternatives with their selecting conditions, including that paediatric_warning_sign_check is faster and uses no AI.

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 declare readOnly/idempotent/non-destructive, and the description goes well beyond them by disclosing determinism ('No model: the same result every time'), the closed rule set, the four possible output levels including the fallback 'routine when no rule fires', and that each firing rule carries its source sheet. That is substantive behavioral context an agent cannot get from annotations alone.

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 tight paragraphs, front-loaded with scope followed by routing guidance; every element earns its place. Slightly dense prose, but no filler or repetition.

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 an output schema exists, the description needn't describe return values, and it still sketches them (level, fired rules with source, warning text, emergency number). Combined with the routing guidance and determinism note, an agent has everything needed to select and call the tool correctly.

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 schema already documents lang, country (including ISO 3166-1 alpha-2 format) and the symptoms field with the age hint. The description's mention of 'eight languages and four scripts' and 'the country's emergency number' restates rather than extends the schema, so the 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 states a precise verb and resource ('checks a description of a child's symptoms against 96 fixed warning-sign rules'), names the rule coverage, the eight languages/four scripts, and the exact symptom domains covered. It is instantly distinguishable from siblings like paediatric_question_with_sources or oral_rehydration_plan.

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 trigger ('Use when symptoms are described, first, to know if the child needs emergency care...'), an explicit ordering ('first'), and routes the agent to two named alternatives with their own conditions (full explanation afterwards; fluids-only cases). Nothing 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.