french-rental-compliance

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

Annotations only carry readOnlyHint/openWorldHint; the description adds substantial behavioral context: rupture conventionnelle is unavailable (CASF L.423-2), the annex-4 retirement indemnity is PAID BY THE INSURER not the employer, formulas are deterministic with no AI assessment, and with no date seniority defaults to today (Europe/Paris) with the result stating so.

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 the purpose well, but the opening sentence is duplicated in French and English, adding redundant length, and the run-on clauses about legal articles are dense. It earns most of its space but wastes some on bilingual 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?

For a 22-parameter, rule-heavy tool with no output schema and thin annotations, the description covers the regime, the reason categories, the key thresholds and the date-defaulting behavior. Return shape is not explained, but with deterministic output that is a minor 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 each of the 22 parameters is already documented in the schema. The description supplies legal rules (art. 120 notice tiers, 1/80 art. 121.1, 9-month threshold) but little parameter syntax beyond what the schema already states, 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?

States a specific verb and resource (end of an assistant maternel contract) with the exact reasons handled (retrait d'enfant, démission, préavis, indemnité, Pajemploi). It explicitly distinguishes itself from calculate_home_employment_end_of_contract (same engine, regime pre-set) and from the garde d'enfant à domicile case.

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 (parent-employer or assistant maternel), when-not ('NOT for a garde d'enfant à domicile — that is regime garde_domicile on the generic tool'), and names the sibling alternative. 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.

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

Annotations already declare readOnlyHint=true and openWorldHint=false, and the description adds substantial non-obvious behavior on top: deterministic rules not AI, 'nothing is assumed' with explicit defaults, disclosure that the fallback grille is ILLUSTRATIVE and that no official grille exists, and that the response flags an assumed date (as_of.assumed). This is exactly the kind of context annotations cannot carry.

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 critical facts are front-loaded in the first sentence, followed by a denser explanatory paragraph. It is long, and the French-language restatement duplicates the English summary rather than adding information for an agent, which costs a point against strict conciseness.

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 hint at returns, and it does: it names what changes by perspective (next_steps), that the result states the legal maximum, and that an assumed date is surfaced in as_of. For a 13-parameter legal calculator this leaves nothing material an agent needs unstated.

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 100% schema description coverage the baseline is 3, and the description earns above it by adding input-shaping guidance the schema lacks: how to phrase damage items ('what, repair cost, age of the element') and what is deliberately omitted when not supplied (lease type, provision amount). It does not, however, add per-parameter meaning beyond what the schema already documents.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

The opening line names a specific verb-resource pair (deposit restitution) and enumerates the three outputs it produces: due date, late penalty, vétusté-apportioned retentions. It is trivially distinguishable from siblings like calculate_rent_revision or calculate_tenant_notice_period without opening any 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?

The description gives clear operating context: it is for a French residential lease, tells the agent which perspective to pass for landlord vs tenant, and instructs how to elicit inputs ('ask the user for amounts and dates in plain words'). It stops short of naming alternative tools or explicit when-not-to-use conditions, so it is strong but not fully routing.

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 openWorldHint=false, so the safety profile is covered; the description goes further by promising deterministic formulas ('no AI, no assessment of any person'), stating that seniority defaults to today (Europe/Paris) when no date is given, that figures are identical for both parties, and that certain contest windows show only their earliest end. These are meaningful behavioral disclosures beyond the 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?

After a serviceable bilingual opening sentence, the body becomes a dense wall of semicolon-chained clauses enumerating rules, article numbers and edge cases. It is thorough but not front-loaded or scannable, and much of the legal detail duplicates the 100%-covered schema descriptions.

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 28-parameter, high-complexity legal calculator with no output schema, the description covers rules, defaults, blocked cases, payer distinctions and input prerequisites. An agent has enough context to invoke correctly and to anticipate the result shape.

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% across 28 params, so the schema already carries the semantic load and baseline is 3. The description mirrors much of that content (notice/indemnity rules, defaults) rather than adding new per-parameter meaning, so no credit above 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 precise verb+resource: computes the end-of-contract figures (notice, dates, indemnity, documents) for a French home-employment contract under CCN 3239, for either party. It explicitly flags the sibling boundary by naming calculate_assistant_maternel_end_of_contract within the regime parameter, so an agent can route between the two.

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 rich conditional guidance on which end_reason/regime combos apply and routes assistant-maternel cases to the sibling tool ('or use calculate_assistant_maternel_end_of_contract'). It lacks an explicit 'do not use this when…' framing, but the routing and prerequisite conditions (e.g. cdd_term_end needs contract_type 'cdd') are clear enough.

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 openWorldHint=false. The description adds valuable behavioral context: it is deterministic Swiss payroll, no AI, and returns annual + monthly net plus deduction breakdown and data year. It doesn't mention rate limits, but for an offline calculation those are less relevant.

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 purpose in French and English, then details deductions. It is concise and every sentence adds information. The bilingual opening is slightly redundant but appropriate for a Swiss tool.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity, the description covers what it computes, the deduction categories, the return format (annual + monthly net, breakdown, data year), and the deterministic nature. With no output schema, this level of return-value detail is exactly what is needed.

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. The description adds that age affects the LPP rate and that canton is the border canton of work, which reinforces the schema's own descriptions. No new syntax or format details are provided beyond the schema, but the added context on age/LPP linkage is marginally useful.

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?

Specific verb+resource: converts gross annual Swiss salary (CHF) to net for a cross-border worker, and names the exact deductions (AVS/AI/APG, AC, LPP, NBU). This is clearly distinguished from siblings like calculate_teletravail_frontalier and compare_lamal_cmu, which handle different payroll/insurance concerns.

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 (convert gross annual Swiss salary to net for a frontalier in a supported border canton), but it does not explicitly state when not to use this tool, such as for non-border workers or cantons outside the enum. The scope is clear enough for most routing decisions.

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 only cover read-only/local execution, yet the description goes well beyond them: it discloses the one-year waiver rule that produces a warning when revision_date is over a year old, the non-retroactive application of the new rent, the F/G exclusion, and that the output is deterministic INSEE data with no AI. This is exactly the kind of behavioral context an agent needs for a legal calculation.

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 key result and the F/G exclusion are front-loaded, which is good, but the opening line is duplicated in French and English, and the body is dense with repeated legal citations (loi n° 89-462, loi n° 2021-1104). It is informative but verbose for a tool 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 4-parameter tool with no output schema, the description is nearly complete: it enumerates the returned value (new rent) and the warning conditions, and covers the two legal limits. It is slightly thin on the exact shape of the result, but an agent has enough to call and interpret it 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 already 100%, so the baseline is 3, but the description adds real meaning beyond the schema: it explains that reference_quarter names the quarter of the revision year and that revision_date wins if they disagree, plus the relationship between the two (a mismatch triggers a warning). Useful interpretation guidance on top of an already-documented 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+resource (computes the legally revised French residential rent) and gives the exact formula (IRL of reference quarter ÷ same quarter one year earlier), naming the governing statute. It implicitly distinguishes itself from siblings like check_dpe_rental_restrictions and generate_rent_receipt by being a calculation 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?

Clear context for when to use it: the yearly IRL revision, with the note that class F/G dwellings are excluded and that dpe_class enables that check. However, it never explicitly routes the agent to sibling tools (e.g. check_dpe_rental_restrictions) for the DPE question, leaving the boundary between overlapping tools 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 establish readOnlyHint=true and openWorldHint=false, so the safety profile is covered. The description adds meaningful behavioral context: deterministic output, official sourced facts with dates, nothing stored, and the distinction between projection and a signed-in real-position log. It does not describe authentication or rate limits, but those are not central for this read-only calculation.

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 very long, dense, and bilingual, with the French translation doubling content without adding new information. It also delays the core action — projecting the telework position — until after a compact block of legal thresholds, so it is not well front-loaded.

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 complex multi-jurisdiction calculation with no output schema, the description does the heavy lifting: it explains what is returned, the canton-dependent meaning of the 40% line, the separate tax and social-security instruments, the mission-day allowance, and the optional option-window logic. An agent has enough context to invoke it correctly despite the lack of a formal output schema.

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 seven parameters are already documented in the schema. The description reinforces how work_canton changes the tax rule and how mission_days_per_year is capped at 10 inside the allowance, but it adds little syntax or interpretation beyond what the schema already states.

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 the specific resource (frontalier telework thresholds under FR–CH rules) and states the operation (projects a typical working week and returns telework share, limit positions, addable days, and mission-day treatment). It is clear enough to distinguish from generic calculators, but it does not explicitly contrast itself with the most plausible sibling, compare_lamal_cmu, even though the LAMal/CMU option window is part of its scope.

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 provides clear usage context: when work_canton should be supplied, what happens if it is omitted, and when the optional LAMal/CMU window applies. It also references the signed-in monthly log path and says list_deadlines then carries threshold warnings, which is useful alternative routing, but it stops short of explicit when-not-use 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 the description reinforces this by labeling the computation 'Deterministic legal formula — no AI involved.' It adds valuable behavioral context beyond annotations: the reception-date emphasis (never the sending date), the automatic zone-tendue resolution, and the return of justificatifs. This exceeds the baseline for tools with annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is dense but well-organized, using a semicolon-separated enumeration and clear cause-effect statements. It front-loads the purpose and then provides necessary legal details. The bilingual intro is slightly redundant, but every sentence conveys essential information, keeping it appropriately concise for the complexity.

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 must convey what the function returns, and it does: 'the end date computed ... plus which justificatifs must accompany the letter.' It also covers the legal grounding, the zone-tendue resolution, and the scope (France métropolitaine). It lacks explicit error-handling notes (e.g., what happens with invalid dates), but for a deterministic calculator the description is sufficiently complete.

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 schema covers 100% of parameters, so the baseline is 3. The description adds significant semantic value: it explains how lease_type dictates the base period, how ground_id and zone_tendue interact to reduce it, and clarifies the crucial nuance of reception_date (landlord receives, not sent). This goes well beyond the schema's raw field descriptions.

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 computes the tenant's notice period for terminating a French residential lease, with specific rules (furnished = 1 month, unfurnished = 3 months with reductions). It names the legal reference (loi n° 89-462 art. 15) and distinguishes itself from sibling tools (e.g., calculate_rent_revision, check_zone_tendue) by its explicit scope and outputs (end date and justificatifs).

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 indicates when to use the tool: for calculating tenant notice periods in France métropolitaine, and even notes it automatically resolves zone tendue, avoiding the need for separate calls. However, it does not explicitly name alternative tools or state when NOT to use it (e.g., for other legal jurisdictions or other lease types), which would push it to 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 state readOnlyHint=true and openWorldHint=false, covering safety. The description adds substantial behavioral context: deterministic rules, legal references (loi n° 2021-1104, loi n° 89-462 art. 17-1), exact threshold dates for bans and rent freezes, and that no AI is involved. It does not mention rate limits or side effects, but for a pure lookup this is strong 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 purpose statement is front-loaded and followed by legal detail that earns its place for a compliance-oriented tool. The bilingual header partially duplicates the title, adding some redundancy, but the overall structure is efficient and not bloated.

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 two-parameter deterministic lookup with no output schema, the description explains exactly what is returned (letting ban status and date, rent freeze, DPE validity) and the condition under which validity is checked. With annotations covering read-only and closed-world behavior, nothing essential 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% and both parameters are documented in the schema itself. The description reinforces meaning (DPE class A–G, dpe_issue_date enabling validity) but adds no syntax, format, or constraint beyond what the schema already provides, 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?

Names a specific resource ('French rental restrictions') and action ('returns') tied to a DPE class. The domain is clearly distinct from sibling calculators like calculate_rent_revision or calculate_tenant_notice_period, though it does not explicitly name or differentiate against those 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 'Given a dwelling's DPE class (A–G), returns...' and the conditional mention that dpe_issue_date enables the validity check. However, there is no explicit when-to-use, when-not-to-use, or alternative-tool guidance, leaving 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 declare a safe read-only, closed-world look-up, so the bar is lower; the description nonetheless adds real behavior — the annual 31 March deadline, non-withdrawability once filed, the legal basis (LIFD art. 99a, OIS art. 14) and the fact that it returns eligibility/share/threshold/deadline/data year. It does not discuss error or edge-case handling, 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?

The core rule is front-loaded and precise, but the bilingual duplication (English plus French) and the long parenthetical on the cantons of the 1983 agreement make it heavier than necessary for a four-parameter look-up.

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 fills the gap by enumerating what is returned (eligibility, share, threshold, deadline, data year) and covers the deadline, irreversibility and territorial scope an agent needs to advise correctly. Nothing material to calling or interpreting 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 all four parameters are already documented in the schema, including the canton/1983-agreement interplay for work_canton and source_taxed. The description restates the 90% rule and the same-currency household framing but adds no syntax or format detail beyond the schema — the 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?

Names a specific rule-check verb and subject (quasi-resident status, ≥90% of household income taxable in CH) and clearly marks itself as a deterministic threshold rule rather than a calculator, which separates it from the sibling calculate_* 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?

States exactly when it applies (person taxed at source in Switzerland living in France, Geneva and cantons outside the 1983 agreement) and when it does not (the eight cantons of the 11 April 1983 agreement, which tax in France). The when-not condition is explicit, not inferred.

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?

Beyond the readOnlyHint=true and openWorldHint=false annotations, the description adds valuable behavioral context: it is a 'deterministic dataset lookup on DILA reference data (service-public.fr simulator dataset) — no AI involved.' It also discloses the real-world consequences of each list, making the tool's behavior and semantic boundary much clearer.

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 information-dense and front-loaded with the core question, then the critical caveat about the two lists. There is minor redundancy from the bilingual repetition of the opening question, but overall every substantive piece of the explanation earns its place given the tool's complexity.

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 absence of an output schema, the description adequately conveys what the caller should expect conceptually: membership in the zone tendue perimeter and which list applies. It also covers the data source, determinism, and input disambiguation needs. It does not specify the exact output shape, but for a boolean/list lookup the semantic outcome is clear enough.

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 the commune name, INSEE code preference, and postal-code disambiguation. The description essentially restates this guidance ('Provide insee_code, or commune (+ code_postal to disambiguate)') without adding new parameter-level meaning 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?

The description uses a specific verb and resource: 'Checks whether a French commune is in the zone tendue perimeter' and immediately distinguishes the two lists (reduced notice vs. tax-only). This clearly separates the tool from sibling tools and makes its exact scope unambiguous.

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 clear usage context by explaining the critical distinction between the two 'zone tendue' lists and warns that the wider TLV perimeter is often mislabelled. It also tells the caller how to provide input ('Provide insee_code, or commune (+ code_postal to disambiguate)'), though it does not explicitly compare against sibling tools or state when not to use it.

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 readOnlyHint=true and openWorldHint=false, so the safety profile is covered. The description adds genuinely useful behavioral context beyond that: deterministic official tables with no AI, named sources (BSV 2026, LAFam art. 3, service-public F13213, Reg. 883/2004 art. 68), and the data year. It does not cover edge cases like unsupported canton/child combinations.

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 core comparison sentence is front-loaded and dense with useful specifics, but the leading bilingual title/duplicate ('Family allowances: Swiss vs French... · Allocations familiales : Suisse ou France...') repeats the same information in two languages without adding meaning, and the single long sentence packs legal citations that a selecting agent does not need.

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 takes on the return shape: Swiss monthly + annual in CHF, CAF monthly in EUR, the differential, per-child detail, notes and data year. Combined with the source citations and determinism claim, this is nearly sufficient for a moderately complex calculator; only unsupported-case behavior is unstated.

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%, with enums and per-field descriptions for canton, children, caf_income_band and spouse_works_in_france, so the baseline is 3. The description adds the legal rationale behind the differential (CAF pays first, Switzerland pays the difference) but largely restates semantics already present in 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 precise verb ('compares') and a scoped resource: family allowances for a cross-border worker under Swiss canton rules vs French CAF, plus the EU 883/2004 art. 68 differential. This is clearly distinguishable from siblings such as compare_lamal_cmu or calculate_teletravail_frontalier.

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 the use case (a frontalier household with children and possibly a spouse active in France) but never states when to use this tool versus alternatives, nor does it name any sibling. The conditions for the differential are described, but only as domain logic, not as guidance on 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 only declare readOnlyHint/openWorldHint, but the description adds rich behavior: values come from deterministic official premium/rate tables (no AI), it recommends neither system, and it returns option rules (window, reception by Swiss authority, default, who holds it) plus form-transmission guidance. This goes far beyond the annotation coverage.

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?

Dense but front-loaded and every sentence carries meaning; the method, fallback, and legal basis are stated efficiently. The duplicated bilingual header line ('Couverture santé frontalier...') is mildly redundant for an agent but does not dilute the core 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?

With no output schema, the description correctly compensates by describing the returns (both monthly figures, option rules, transmission steps). All five parameters are documented at full coverage, so an agent has everything needed to call it 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 beyond the schema: income_n2_eur is the worker's OWN net N-2 income, explicitly 'not the household revenu fiscal de référence', and dependants are priced by age band. salary_chf's fallback role is also clarified.

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 (estimates monthly cost) and a precise resource (the two frontalier health-insurance options, LAMal vs CMU/PUMa), and frames it as the droit d'option choice. An agent can distinguish this from sibling calculators like compare_family_allowances purely from the text.

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 conditional use of parameters: give income_n2_eur for the exact French figure, otherwise salary_chf yields an upper estimate explicitly flagged as such. This covers when and how to invoke it well, but names no alternative sibling tools or exclusion conditions.

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 readOnlyHint=false and idempotentHint=false, and the description supplies the missing behavioral detail: the PDF lives 24 hours behind a single-use link then is deleted, per-network rate caps return a structured rate_limited result with retry guidance, missing mentions print blank and are never invented, and refusals persist even with acknowledge_incomplete. That is far beyond what the annotations state.

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 critical information (free, French-only output, blank mentions, needs_input flow) is front-loaded, but the description is very long and its account/membership paragraph is dense commercial detail (€4.90, €6/month, €59/year, 10 properties, extra-property rates, e-signature) that contributes little to selecting or correctly invoking the tool.

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 14-parameter tool with no output schema, the description closes the important gaps: it names the interactive return states the agent must handle (needs_input/blank_mentions, printed_blank, rate_limited), the 24-hour link lifetime, and the mandatory-French constraint. No output schema exists, so the return-value explanation it does provide is warranted.

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 already 100%, so baseline would be 3; the description nonetheless adds cross-parameter semantics the schema cannot express — omitted mentions become printed blanks, includeEnglish only appends a non-binding translation while the document stays French, and acknowledge_incomplete is meaningful only after a needs_input response. It does not add format examples beyond the schema's own patterns.

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 artifact: produce a French rent receipt (quittance de loyer) PDF from supplied landlord/tenant/period/rent data. The legal frame (loi n° 89-462 art. 21) and 'deterministic template engine, not AI-generated' pin down exactly what is produced, which is distinguishable from the calculation-only siblings that merely compute amounts or eligibility.

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 operating conditions: have the landlord verify every field, payment happens in the browser never through the agent, and the acknowledge_incomplete flag must never be set on a first call — only after a needs_input response has been shown to the landlord. It also describes the account-based variant and pricing path, so an agent knows both the free flow and when the paid one applies.

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