לצדך: Israeli health insurance, from the policies

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

Annotations already declare the operation read-only, idempotent, and non-destructive, so the description does not need to cover side effects. It adds useful detail about the return contents (full content, sources, page URL), though it does not disclose error behavior or edge cases.

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 English description is compact and front-loaded with the core purpose. However, the full Hebrew translation is redundant for an AI agent consuming this definition, so not every part of the description earns its place.

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 fetch tool with complete schema coverage, the description is largely sufficient: it explains what the output contains and what valid ids look like. It could benefit from a short note on invalid-id behavior, but nothing essential is missing for basic 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 parameter is already described as 'An id returned by search.' The description adds meaningful value beyond the schema by enumerating valid id shapes like 'question:<slug>', 'test:mri', and 'surgeon:<licence>', which helps the agent construct correct inputs.

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 uses a specific verb-resource pair: it fetches the full content of one search result by id, and lists concrete id formats. It is clear, but it does not explicitly differentiate itself from siblings like get_cover or search, leaving some differentiation 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?

The phrase 'id from search' implies this tool is intended to be used after a search returns an id, which is useful contextual guidance. However, it never states when not to use it or names alternatives, so the usage guidance remains implicit rather than explicit.

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 explicitly states an open-world behavior: absence from a list does not mean no arrangement, and some lists are not checked yet. This directly contradicts the annotation openWorldHint=false, which indicates that missing results are conclusive. Because the description contradicts its own annotations, it scores the minimum.

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 English sentence front-loads the purpose and criteria efficiently, and the caveat is important. However, the Hebrew duplicate repeats the same content, so not every sentence earns its place.

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 lookup with no output schema, the description explains what will be returned (lists with dates), the critical absence caveat, and the required user verification before surgery. It does not explain limit behavior or multi-match handling, but those are minor gaps given the annotations.

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 80%, so the schema already documents most parameters. The description adds only the 'forgiving spelling' nuance for name and otherwise re-states the search criteria; the limit parameter receives no added meaning.

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 a specific action – find a surgeon by name, licence, specialty, or region – and the output: which arrangement lists and dates. The resource and outcome are unambiguous, though it does not explicitly contrast with siblings such as get_cover or search, so full differentiation credit is not given.

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 this is the tool for surgeon arrangement lookups but never says when to prefer it over the sibling tools or what conditions rule it out. The caveats about list coverage and insurer confirmation are domain advice, not tool-selection guidance.

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 readOnlyHint=true and idempotentHint=true, so the description does not need to repeat safety traits. It adds valuable context: facts are quoted from the insurer's published document, and the output includes what the document does not say and premiums as printed. This goes beyond annotations to explain the tool's output nature.

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 English description is concise and front-loaded with the core purpose and output details. The Hebrew translation is redundant but adds no extra info. The description earns its place without unnecessary 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?

For a read-only retrieval tool with no output schema, the description thoroughly explains what the agent will receive: summary, quotes with page/section, what's not said, common questions, and premiums. It covers the essential expectations for calling the tool correctly. No missing critical info.

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%: both parameters have descriptions, including accepted formats for cover and a list for company. The description adds a concrete example (company 'הראל', cover 'תרופות מחוץ לסל'), clarifying how to specify the parameters beyond the schema alone.

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: get one cover of one insurer, and enumerates the output (plain summary, key facts with quotes and page/section, what the document does not say, common questions, monthly base premiums). This clearly differentiates it from siblings like list_companies or search, which list or find items rather than detail a single cover.

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 for retrieving details of a specific cover, and the parameter description references list_companies, but there is no explicit when-to-use vs alternatives, no exclusions, and no mention of when not to use this tool. Usage guidance is only implicit.

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 readOnlyHint, idempotentHint, and no destructive operations, covering safety. The description adds valuable behavioral context beyond annotations: the list is alphabetical, not ranked, limited to 7 insurers 'we read' (implying a curated static set), and includes page URLs and summary existence flags. This clarifies output characteristics and scope, which annotations do not convey.

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 sentences, the first English and the second a Hebrew translation. The English sentence is information-dense and front-loaded: it names the exact resource set, ordering, ranking, and output fields. No redundant words; the Hebrew duplication is arguably unnecessary but standard for bilingual audiences, not a structural flaw. It earns its place.

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 no output schema and no parameters, the description is complete for an agent to call correctly: it states what the list contains (7 insurers, their covers, URLs, summary flags), the ordering, and the selection criteria ('we read'). There is no additional context an agent needs to invoke this tool successfully. The sibling tools are contextually distinct, and the description fully covers the tool's behavior.

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 has zero parameters, so the schema is fully covered by absence. Per calibration, with 0 params, the baseline is 4. The description adds no parameter details because none exist; it instead clarifies output semantics, which 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 ('list') and resource ('insurers and their health covers'), and adds precise scope: the 7 Israeli insurers read, alphabetically with no ranking, including page URL and plain-language summary existence. It clearly distinguishes from siblings like search or get_cover by being an exhaustive overview of a fixed set.

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: it lists what has been read, so an agent can infer it's the go-to for an overview. However, it does not explicitly state when to use it versus alternatives (e.g., 'use get_cover for details' or 'use search for dynamic queries'). No exclusions or when-not-to-use guidance is provided.

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 readOnlyHint=true, idempotentHint=true and destructiveHint=false, so the tool's non-mutating safety profile is fully covered by structured data. The description adds one genuinely useful behavioral trait — outputs are routing keys/identifiers consumed by other routes rather than free-form data — which clarifies the nature of the return values. No contradiction with annotations; moderate added value.

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 English sentence is compact and front-loaded, packing the topic list and key-routing hints efficiently. However, the trailing Hebrew line ('כל הנושאים: בדיקות, רופאים מומחים, שאלות, סוגי כיסוי') is redundant with the English content and is actually incomplete — it omits refund services, surgeon specialties and regions — adding confusion rather than value. That redundant, slightly inaccurate sentence costs the tool a higher score.

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 zero parameters and rich safety annotations, the description covers the main need: what topics exist and which sibling routes consume their keys. But there is no output schema, and the description never states the response shape (e.g. flat list of strings vs key-value pairs, or whether region/cover values come grouped), leaving the agent to guess the return format. Adequate but with a notable gap for a no-schema 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 tool takes zero parameters and additionalProperties is false, so there is nothing for the description to document. The 0-param case is the easy baseline of 4, and the description appropriately focuses on what the response contains rather than inputs. No parameter ambiguity exists to resolve.

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 clear verb-plus-resource purpose: listing all topics the site answers, enumerating categories (tests, specialties, question pages, cover kinds, refund services, surgeon specialties/regions). It goes beyond the title by mapping each topic to the sibling tool that consumes its keys, distinguishing it from siblings like find_surgeon, refund_estimate and test_fast_routes. It falls short of a 5 because it never explicitly names itself as the enumeration/prerequisite tool.

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 guidance is conveyed implicitly through the key-mapping parentheticals — e.g. tests produce keys for test_fast_routes, refund services for refund_estimate — which lets an agent infer 'call this to get valid keys before using those tools'. However, there is no explicit when-to-use/when-not-to-use statement, no mention of alternatives, and no context on which sibling to fall back on. Useful but 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?

Annotations already declare readOnlyHint, idempotentHint, and non-destructive behavior. The description adds behavioral nuance beyond annotations: it discloses that ambiguous documents yield both readings and that missing refund numbers are explained. It also warns that only checked covers are included and that the person's own policy is authoritative. No contradiction with annotations was found.

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 English portion is front-loaded and efficient, but the Hebrew paragraph duplicates the same content without adding new information for an agent. This redundancy prevents a higher score, though the structure remains readable and the key usage context appears early.

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 carries the burden of explaining what the tool returns. It does so well: per insurer/cover, both numbers when the document is ambiguous, and reasons when no number is present. The main gap is not explicitly describing the optional company parameter's behavior, but the schema already covers that.

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 input schema already describes all four parameters. The description adds light context by tying service to 'private test or specialist visit' and amount_paid to the refund calculation, but it does not need to repeat schema details. This meets the baseline for fully documented schemas.

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 names the operation: for a private test or specialist visit and an amount paid, it reports what the documents allow saying about a refund, per insurer and cover. It also describes how ambiguity is handled ('both numbers') and why a number may be absent. This distinguishes it from generic siblings like search or get_cover, though it does not explicitly name the sibling it is not.

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 when to use the tool: when there is a private test or specialist visit and an amount paid. It also states a scope limitation: only covers that were checked, and the person's own policy decides. However, it does not explicitly mention alternative tools such as get_cover or search, leaving some routing 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?

The description discloses real-world side effects beyond what annotations convey: a licensed agent will contact the person by phone or WhatsApp and may offer insurance. It also imposes a strict privacy constraint about omitting medical details. This is meaningful behavioral context on top of readOnlyHint=false and idempotentHint=false.

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 English portion is front-loaded and efficient: action first, then consent and privacy constraints. The Hebrew repetition adds bulk without new information, so it is not perfectly concise, but the structure is still clear and scannable.

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 handoff tool with 7 parameters and no output schema, the description covers purpose, consent gating, privacy limits, and the real-world callback behavior. It does not describe the return value or post-submission flow, so a small completeness gap remains, but nothing essential for safe invocation 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 baseline is already 3. The description reinforces that user_consent must be true and that medical details should not be included, but it does not add new parameter semantics beyond what the schema already states for each field.

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 uses a specific verb and resource: it "hands the person over to a licensed insurance agent," who calls back to check the person's policy. This clearly differentiates it from sibling tools, which are all information-retrieval or estimation operations rather than human handoffs.

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 explicit conditions: call only after the person explicitly asked and consented, never on your own initiative, never with details the person did not provide, and only with user_consent=true. This is direct when-to-use and when-not-to-use guidance with no ambiguity.

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 (readOnlyHint, idempotentHint, destructiveHint all signal a safe read). The description adds value beyond those by disclosing the result contract: it returns lightweight pointers ('ids, titles and URLs') rather than full facts, and scopes what is searched. 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 English portion is dense, front-loaded with the verb and scope, and uses vivid real-user examples. However, the trailing Hebrew sentence is a near-total duplicate of the entire English description rather than a compact gloss, so it does not fully earn its place in an agent-facing definition.

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 correctly takes on the job of explaining the return shape ('Returns ids, titles and URLs') and next-step routing to fetch/specific tools. Gaps are minor: limit semantics, result ordering, and no-results behavior are not mentioned, but for a two-parameter read-only search tool this is mostly 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 coverage is 50%: query is well-documented with examples in the schema, and the description reinforces the free-text/bilingual semantics and content categories. However, the limit parameter has no description in either place, and the description does not confirm that limit caps the number of returned ids/titles/URLs. At 50% coverage, the description should compensate more.

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 ('Free-text search... across the site') and enumerates the exact content scoped: people's questions, tests, specialists, covers of 7 insurers, and surgeons on arrangement lists. This clearly distinguishes it from siblings like find_surgeon or get_cover, which target narrower slices of the same domain.

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 clear context that search is the discovery entry point and explicitly routes onward: 'use fetch(id) or the specific tools for the full facts.' This tells the agent when search is insufficient, though it stops short of naming conditions for preferring specific siblings (e.g., 'use find_surgeon when you already know the surgeon's name').

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 readOnly and idempotent behavior; the description adds useful context such as dated figures, exact word-for-word promised times, what the patient pays, and whether the document names the specialty. It also flags the psychiatry exception, going beyond what the annotations convey.

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 English portion is information-dense and front-loaded with the specialty scope, but the Hebrew sentence repeats the same content almost verbatim, adding length without new information. This duplication prevents a higher score.

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 lookup with only two parameters and no output schema, the description is largely complete: it states the data sources, the per-insurer breakdown, the refund rows, and the psychiatry special case. It does not describe formatting, but that is a minor gap for 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?

Schema description coverage is 100%, so the baseline is 3. The description restates the specialty scope and per-insurer behavior but adds no new parameter-level detail beyond the schema, which already documents the allowed specialty list and company behavior.

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's resource: one specialty's Ministry wait figures, per-insurer online-consultation and fast-track routes, and private-visit refund rows. It is specific enough that an agent can tell it from broad tools like fetch or search, but it does not explicitly name a sibling it is not, so it stops short of full differentiation.

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 opening 'For one specialty' gives a clear selection criterion, and the psychiatry note adds a special-case rule. However, it does not explicitly say when to prefer this over siblings like get_cover, find_surgeon, or refund_estimate, leaving some routing 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 mark the tool as read-only and idempotent, and the description adds substantial behavioral detail: only official dated figures are used, 'otherwise says there is none', promised times are quoted word for word, and document gaps are disclosed. This goes well beyond what the annotations alone provide.

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 English description is front-loaded with scope and enumerates return content in a compact, colon-separated structure. The Hebrew translation is redundant for an AI-facing definition and slightly lengthens the text, but the organization remains clear.

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 carries the burden of explaining what the agent should expect, and it does so comprehensively: kupa wait figures, insurer fast-track specifics, cost stages, waiting periods, and open questions. Combined with the detailed parameter schema, an agent has enough context to invoke 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 explains both 'test' and 'company' parameters, including the Hebrew/English options and the behavior when company is omitted. The description mostly restates this context rather than adding new parameter-level meaning.

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 handles one medical test at a time and reports kupa wait times plus every insurer's fast-track route, including who decides, how to start, promised times, costs, and waiting periods. This resource ('test fast-track routes') is distinct from siblings like 'specialist_routes' or 'find_surgeon'.

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: use it for one test, with optional filtering to a single insurer via the company parameter. It does not explicitly name alternatives or say when not to use it, but the scope is unambiguous enough to guide selection among siblings.

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