IUSTORIA – průvodci a právní kalkulačky

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

Annotations already cover the safety profile (readOnlyHint=true, idempotentHint=true, destructiveHint=false). The description goes far beyond them: it discloses the exact rate rules, date-dependent fuel price changes in 2026, trailer percentage, lost-time rate of 150 CZK per half-hour, VAT handling, and the years covered. This is rich behavioral context for a pure computation tool.

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

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

The core calculation is front-loaded in the first clause, and the description is not padded. However, it is delivered as a very long, dense Czech sentence followed by a brief English summary; for a 10-parameter legal calculator this length is justified, but structure could be improved with clearer separation of rate-period details.

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 and lack of an output schema, the description should ideally describe what the computed result looks like (e.g., breakdown of base compensation, fuel, lost time, VAT). It does not mention return values or error behavior at all, leaving an agent unsure about the output shape despite otherwise thorough calculation rules.

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 repeats most param semantics already in the schema (prives +15%, platce_dph 21%, cena_z_dokladu vs decree average). It adds only the specific 150 CZK lost-time rate, which is marginal new param meaning beyond what the schema provides.

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 ('Spočítá náhradu') and two resources (travel costs by own car, lost time compensation) with exact legal references. The word 'advokáta' distinguishes it from the employee-travel sibling cestovni_nahrady_zamestnance. An agent can tell what it computes without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives clear usage context: for Czech lawyer journeys by own car, with verified rates only for years 2023–2026. It also states what it does not calculate (stravné, delay of hearing), which is useful when-not guidance. However, it never names the sibling tool cestovni_nahrady_zamestnance, so explicit alternative routing is missing.

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?

Goes well beyond the readOnly/idempotent annotations: it discloses input caps (30 l / 40 kWh, 100 Kč/l, 30 Kč/kWh), a warning trigger (average speed over 130 km/h), the abort behavior on days without published rates, and the year-specific rate tables driving 2024–2026 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?

Front-loaded with the core function, then layered rate/limit detail. It is dense and long, but nearly every clause is load-bearing legal/boundary information rather than filler; the trailing EN summary aids the agent.

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 nested-parameter, no-output-schema calculator the description covers computation rules, caps, warnings, exclusions, and edge-case behavior (missing year rates). Only the shape of the returned result is left undocumented, which is a minor gap given the absence of an 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 coverage is already 100%, so baseline is 3; the description adds real semantic value by explaining what each input drives (sfera → dolní sazba rozpětí, krácení percentages 70/35/25, foreign rate tiers over 18/12/1 h, fuel price valid on travel date). It stops short of documenting the nested vozidlo defaults beyond what the schema 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?

States a precise verb+resource ('Spočítá cestovní náhrady zaměstnance') and immediately scopes it to the components computed: domestic stravné, foreign stravné, and private-vehicle mileage. This clearly distinguishes it from sibling tools like cestovne_advokata or nahrada_mzdy_pri_nemoci.

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?

Makes the use case explicit (employee business trip allowances, with legal citations §163, §170, §157/158) and states exclusions ('Kapesné, ubytování a jízdenky nepočítá'). It does not name an alternative sibling to route against, so it stops short of a 5.

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

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

Annotations already declare readOnly/idempotent/non-destructive/openWorld=false, so safety is covered; the description adds genuinely non-obvious behavior: rounding up to whole CZK, terminating exactly at acquisition cost, refusing years outside 2008–2027, and emitting a warning (not a computation) for § 30a extraordinary depreciation. It does not describe the shape of the returned schedule, but with no output schema that is a minor gap.

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 purpose before the legal detail, and nearly every clause earns its place (method, groups, rounding, year window, threshold, warning). The English restatement roughly doubles the length and is largely redundant for an agent, which is the only real inefficiency.

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 self-contained calculator with no output schema, the description covers computation rules, valid input ranges, rounding, and edge-case behavior (out-of-range years rejected, § 30a only warned about). What is missing is any hint at the result structure — e.g. a per-year table — which the agent must infer.

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 it: it narrows rok_zarazeni to 2008–2027 although the schema permits 1990–2100, ties zvyseni_10 to groups 1–3 and excludes passenger cars, and states the 40,000/80,000 CZK threshold split by year. These constraints are not derivable from 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?

Opens with a concrete verb+resource ('Sestaví daňový odpisový plán hmotného majetku') and pins it to § 30–32, naming both supported methods (rovnoměrně § 31, zrychleně § 32). No sibling tool covers depreciation, and the scope is stated so precisely that the agent needs no schema to identify it.

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 strong context: which assets, which years (2008–2027), which groups, and two explicit exclusions — it does not classify assets into groups ('Majetek do skupiny nezatřiďuje') and does not compute extraordinary depreciation ('které nepočítá'). It never routes to an alternative sibling or says when another tool is the better pick, so it falls short of a 5.

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

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

With annotations already confirming this is a read-only, idempotent, non-destructive calculation, the description adds rich behavioral context: historical rule versions, date-dependent legal regimes, partial-payment handling, deferral-interest handling, and return caveats such as reporting only a partial amount from 2021 onward. It also clearly states limits on what the tool will not do, such as not deciding remission.

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 purpose is front-loaded in the first sentence, which is good. However, the description is an extremely dense wall of legal clauses and citations; while much of it is domain-relevant, it is not tightly structured and could be more scannable for an agent.

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 high complexity, 20 parameters, rich schema, and absence of an output schema, the description covers most selection-relevant behavior: modes, historical limitations, and some return caveats. It could still say more about the returned shape or fields, but overall it is substantially complete for calling 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 coverage is 100%, so the schema already documents all 20 parameters thoroughly. The description adds meaningful legal and interaction context beyond the schema, especially for druh modes, date-dependent rules, konec_lhuty usage, and how partial payments and deferrals affect the calculation.

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 gives a specific verb and resource in Czech and English: it calculates Czech tax penalties under the Tax Code. It further distinguishes its four calculation modes (druh=urok, pokuta, penale, kontrolni_hlaseni), so an agent can tell what the tool does without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description gives clear contextual usage rules for each mode and several explicit when-not boundaries (e.g. it does not calculate for tax due before 1 Jan 2011, does not substitute an unknown CNB rate, and does not decide on remission). It does not, however, explicitly compare itself to sibling tools or name an alternative to use instead.

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

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

Annotations already declare readOnly/idempotent/non-destructive, so the bar is low, but the description adds real behavior: weekend/holiday shifts to a working day with an explanation, past deadlines are reported, and for income tax a still-running deadline is offered. It doesn't describe performance limits or auth, but for a pure calculator those are irrelevant.

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 purpose and organized by tax type, with a useful EN summary at the end. It is dense and long, with a legal citation (čl. II bod 3 zákona č. 283/2020 Sb.) that borders on noise, but for a multi-tax domain tool the length is largely earned.

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?

No output schema exists, yet the description hints at return behavior (reports overdue deadlines, offers a running income-tax deadline) and covers all tax types, period bounds and edge cases. It does not fully specify the response shape, which is the one gap for a 9-parameter tool without an 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% and the enum descriptions already map each code to its meaning, so the schema does the heavy lifting. The description reinforces the substantive rules behind dpfo/dppo/dph/dznv/silnicni and zpusob, 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?

States a specific verb and resource ('Spočítá lhůty pro podání daňového přiznání a splatnost daně') and enumerates the exact tax types and rules it handles, so an agent can distinguish it from siblings like lhuta, promlceni or danove_sankce without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives clear usage context: it covers periods 2020–2030, lists the taxes and the special cases (hospodářský rok before 31.12.2020 excluded), and explicitly bounds scope with 'Nerozhoduje, zda povinnost přiznání máte.' However it never names an alternative sibling tool or an explicit when-not-to-use-this condition.

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 read-only, idempotent, non-destructive, and closed-world behavior. The description adds substantial behavioral context beyond annotations: it explains the exemption rules, taxable threshold logic, reporting deadline for exempt income over CZK 5 million, year-dependent rules, and that it performs no tax 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 front-loaded with the core purpose and key exemption rule, followed by necessary legal conditions and an English summary. It is long and dense, but the detail is mostly justified by the legal complexity and six-parameter surface.

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 legal complexity, rich schema descriptions, annotations covering safety, and no output schema, the description is complete enough for correct invocation. It covers the relevant exemptions, thresholds, year limits, reporting obligation, and explicit exclusions.

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 contains detailed parameter descriptions and enum value meanings. The description adds legal context around thresholds and relationships, but it does not add parameter syntax or usage detail beyond what the schema already provides, so the baseline of 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 (Řekne, zda = says whether) and resource (gift, inheritance, legacy received by an individual), and explicitly distinguishes the tool from tax calculation by noting that it does not compute tax and that Czech gift/inheritance tax no longer exists. An agent can tell this is a Czech income-tax exemption classifier, not a tax calculator.

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 the tool applies: years 2014–2027, gifts, inheritance, and legacies, with thresholds and reporting rules. It also states what the tool does not cover (older rules under Act 357/1992 and tax calculation), but it does not name a sibling alternative for those excluded cases.

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 the read-only/idempotent safety profile, and the description adds substantial behavioral rules beyond that: refusal for pre-2014 dates, non-computation of class 6, the order of operations (halving community property first), treatment of disinherited/renounced descendants, and that the shortfall is returned as a monetary claim under § 1654.

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?

It is a single dense paragraph, front-loaded with the core purpose before drilling into class-by-class rules and limitations. Every sentence carries a legal rule or constraint, though the density is high for a description.

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

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

For an 18-parameter, output-schema-less legal calculator, the description covers what it computes (shares, forced shares, community-property split, monetary shortfall), its exclusions, and its input convention. Nothing essential to calling 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% with bilingual per-field documentation, so the schema already carries the parameter burden. The description only restates the position-based, name-free ordering convention (dite1, dite2 …) that the schema also documents, adding little beyond the 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?

The description names a specific verb and resource (computes statutory inheritance shares and forced shares under the Czech Civil Code) and enumerates the classes and rules covered. An agent can immediately tell it apart from the sibling legal calculators (tax, wage, court fees) which serve entirely different domains.

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 states explicit boundaries: only for deaths from 1 Jan 2014, class 6 is not calculated but flagged, and persons are entered by position rather than name. It does not name alternative tools, but in this sibling set there are no direct alternatives, so the context is clear.

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, idempotent operation. The description adds substantial behavioral detail beyond that: rounding rules, maximum entitlement cap, sick-leave handling, refusal conditions, warnings for overtime and earlier years, and minimum-wage adjustments. 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 main purpose is front-loaded, but the description is a dense, lengthy paragraph followed by a redundant English summary. While most legal detail is relevant, the lack of structure and duplication reduce 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?

Given the tool's complexity (14 optional parameters, no output schema, rich annotations), the description covers legal rules, input scenarios, edge cases, year limits, and refusal conditions thoroughly. An agent has enough information to invoke 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 schema already documents all 14 parameters in detail. The description adds some legal context and cross-parameter relationships (e.g., sick leave can be given as hours or periods), but it largely duplicates schema-level guidance rather than adding new syntactic or semantic information.

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 (Spočítá) and resource (dovolenou v hodinách) and immediately scopes it to Czech statutory law from 2021. The purpose is unambiguous and does not merely restate the title or name.

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?

Provides clear context for when the tool applies (employment relationships, DPP/DPČ, manually entered hours, years 2021–2027) and notes limitations such as refusing DPP over 300 hours and not counting overtime. However, it does not explicitly name alternative sibling tools or state when to use another tool instead.

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

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

Annotations already declare readOnly, idempotent, non-destructive and closed-world, so the safety profile is covered. The description goes well beyond that, disclosing date-dependent rate regimes (21/12% from 2024, historical 10% in 2015–2023, older statutes since 1993), threshold rules, filing deadlines, and that the first result line states the applicable rates.

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 main purpose is front-loaded, which is good, but the body is one extremely dense paragraph replete with historical legal detail (coefficients to four decimals, 1993 rules, 2013–2024 thresholds) that is hard to parse and arguably beyond what an agent needs to select 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 7-parameter, no-output-schema tool this covers both operating modes, all defaults, the date-driven behavior, and an explicit non-coverage caveat, plus a hint about the result's first line. It is close to complete, with only the result shape left somewhat implicit.

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 schema already marks each parameter as 'Jen u vypocet' or 'Jen u registrace' with defaults, and the prose largely restates those same mode/rate/date relationships rather than adding syntax or format detail beyond them.

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

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

The description opens with a precise verb+resource ('Spočítá DPH podle českého zákona o DPH') and then separates the tool's two concrete modes (vypocet vs. registrace) with exact semantics for each. The purpose is unmistakable, though it never names or contrasts a sibling tool, so it clears 4 but not the sibling-differentiation bar for 5.

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 clearly signals when each mode applies ('ucel=vypocet ze základu... nebo z ceny včetně daně', 'ucel=registrace spočítá, odkdy je osoba... plátcem') and adds an explicit scope exclusion: it does not determine whether a supply falls under a reduced rate or is exempt. What is missing is routing to any alternative tool, so it stops just short of a 5.

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

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

Annotations already declare the safe read-only, idempotent, closed-world profile, so the description is not needed to establish safety. It does add valuable behavioral context beyond that: the 2026 vs 2027 tax-regime switch, the derived 2027 thresholds, the pensioner contribution discount, and the explicit caveat that 2027 figures are computed before official announcement. It does not describe the returned result shape, which matters slightly given no output schema.

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 purpose is front-loaded, but the body is very long and dense, packed with statutory citations and exact figures, and then largely restated in an English paragraph. For a legal-rule calculator much of this earns its place, yet the duplication and citation density push it below an efficient, tightly-structured 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 complex, 9-parameter legal calculator the description covers the rule set, year-dependent behavior, and edge cases (health-insurance top-up to the minimum base, minimum-wage shortfall) thoroughly. It never spells out the returned values, and with no output schema that is a minor remaining gap, but nothing essential for correct 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 schema already documents every parameter with Czech and English glosses. The description restates the same rules (300-hour DPP limit, 2026/2027 thresholds) but adds no syntax or format detail the schema lacks, 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 specific verbs (checks limits, determines insurance participation, computes net pay) applied to a precisely named resource (DPP and DPČ agreements). It enumerates exactly which legal checks it performs (§ 75/76 hour limits, § 111 minimum wage, § 7d/7e insurance, tax regime), making it clearly distinguishable from generic siblings.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

The scope is implied by the described calculations, so an agent can infer this is the tool for DPP/DPČ pay and insurance questions. However, it never explicitly states when to prefer this over close siblings such as minimalni_mzda_a_priplatky, nor does it name any alternative or exclusion. No explicit when-to-use or when-not guidance is given.

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 carry the read-only/idempotent profile, yet the description adds substantial behavioral context: which cases are out of scope (pre-2017), which components are deliberately omitted and why (§ 12 odst. 2 double reading), and the fact that it surfaces a upozornění flagging both readings and the § 88 defense route. This is genuine disclosure 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?

Front-loaded: the first clause states exactly what is computed, then enumeration of the components follows. It is dense and runs long, and the closing legal-advice sentences (§ 46/§ 88) drift toward scope creep for a calculator, but nothing is fatally redundant.

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 8-parameter calculator with no output schema, the description covers scope, exclusions, versioning and edge cases thoroughly, and even hints at an upozornění in the result. It stops short of describing the numeric return shape, but with no output schema and exhaustive schema docs that gap is minor.

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 meaning the schema alone doesn't: it explains that den_povereni/zahajeni_exekuce can be supplied instead of the povereni enum and clarifies how the tariff version is selected. It doesn't add value to vymahano/vymozeno thresholds 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?

States a precise verb+resource: it computes a Czech bailiff's fee (odměna) plus flat-rate expenses under vyhláška 330/2001, naming the exact tariff sections and rounding rules. It is clearly distinguishable from siblings like odmena_advokata, naklady_sporu and soudni_poplatky 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?

Gives clear applicability conditions – it explains the drei tariff-version branches (pred2026 vs od2026 vs pred2017) and that pre-1.4.2017 enforcement is not calculated, plus that the § 12 záloha is deliberately excluded. It does not name alternative sibling tools to route to, so it stops short of explicit when/when-not/alternatives.

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 the safety profile (readOnly, idempotent, non-destructive, closed-world), so the description carries the rest and does so richly: max five results, each with section + short answer + link, one-section word-intersection matching, a defined no-results behavior, and the dictionary-term tie-in.

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 purpose and result shape, and every sentence contributes operational detail. The appended EN summary restates much of the Czech text, adding length without new information, which keeps it just below a 5.

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 fully specifies the return shape (up to five pages, each with section, short answer, link), the fallback response, and the scope/language behavior. An agent has everything needed to call it correctly and interpret results.

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 the baseline is 3. The description still adds value by explaining the semantics behind rozsah (services only on explicit request, default guides+blog) and jazyk (guides Czech-only), which clarifies how the enums behave in practice.

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 verb (vyhledá) and resource (stránky průvodců IUSTORIA a blogové články advokátů), and states the retrieval model (hledá po oddílech). It explicitly distinguishes itself from the sibling nacist_stranku, which is named as the tool that returns full page 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?

It gives explicit routing: full text via nacist_stranku, services pages only when rozsah=sluzby/vse and only on explicit user request, and language selection behavior (guides Czech-only, other languages hit the blog). The no-match fallback behavior is also spelled out.

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 safety (readOnly, idempotent, non-destructive). The description adds valuable domain behavior: it states the tool will mark notice in a protected period as prohibited, lists the legal versions and date ranges covered (2007–2027, DPČ from 1993, DPP from Oct 2015), and details historical rules. This goes well 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?

The description is front-loaded with its main purpose, but it is a dense, single-paragraph block in Czech followed by an English translation. While packed with relevant legal detail, it is verbose and could be better structured (e.g., bulleted) for quick agent parsing.

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 thoroughly covers inputs, legal rules, date ranges, and historical versions. However, with no output schema, it does not describe the return format or values, leaving a gap in what an agent should expect back.

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 15 parameters are fully documented in the schema. The description adds little parameter-specific meaning beyond what the schema provides; baseline 3 is appropriate when the schema carries the semantic burden.

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 and resource: 'Spočítá, kdy skončí pracovní poměr po výpovědi...' and enumerates many calculated outputs (notice period, severance, deadlines). It is highly specific, but does not explicitly differentiate itself from siblings like dpp_dpc or podpora_v_nezamestnanosti, even though it overlaps in 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?

The description explains what the tool calculates but gives no guidance on when to use it versus alternatives. There are no exclusions, prerequisites, or named sibling tools to route the agent, leaving selection 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 readOnly/idempotent/non-destructive, so the safety profile is covered. The description adds genuine behavioral context beyond that: it enumerates the counting rules applied, the holiday regime by year, service-by-fiction handling, and the important scope limits (does not compute legal-force date, restricted historical 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?

Front-loaded with the core action, then organized into scope, rules, exclusions, and historical caveats. It is long and citation-dense, but for a legal-calculation tool most sentences carry an operative rule rather than 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?

An 8-parameter tool with no output schema is well covered: inputs are explained via schema, and the description conveys the calculation semantics and limitations an agent needs. The only gap is a clearer statement of the returned breakdown, which is only implied by 'rozepíše výpočet krok po kroku'.

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 itself documents every parameter including defaults and enum meanings. The description reaffirms defaults and rule behavior but adds little parameter-level detail the schema lacks, meriting the baseline 3.

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 ('Spočítá poslední den lhůty a rozepíše výpočet krok po kroku') and scopes it to Czech procedural and substantive deadlines. An agent can distinguish this from siblings like promlceni or lhuty_v_pojisteni by the explicit legal framing.

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 on when it applies (§57 OSŘ procedural, §605-607 OZ substantive, statutory deadlines like odpor/odvolani/dovolani, court-set deadlines) and states exclusions ('Den právní moci nepočítá', no dovolání before 2001, no DS/ulozeni before 2009). It does not explicitly name alternative sibling tools for overlapping questions, so it falls short of a 5.

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

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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive, closed-world), yet the description still adds substantial behavior: refusal of parameters irrelevant to the chosen regime, rejection of unsupported dates, the conservative rule of returning the earlier date where interpretation is uncertain, and the special statutory basis for older contracts. It does not describe the shape of the returned result, 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 purpose and legal basis are front-loaded, which is good, but the single block is extremely dense with parenthetical statute citations (§ 2797, § 2798, § 2805/2806/2807/2786, § 3028 odst. 3) and then largely duplicated in the EN paragraph. Much of the legal-citation detail could be trimmed without losing agent-relevant guidance.

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, multi-regime legal calculator with no output schema, the description supplies the regime logic, era-dependent statute selection, rejection rules, and ambiguity policy, and even states in prose what the tool will tell the user. It omits any indication of the response structure, which for a tool this complex is a minor but real 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% and every parameter carries its own Czech+English description with enum meanings and defaults, so the schema does the heavy lifting. The prose adds legal context (which era drives which law) but little per-parameter meaning beyond the schema. Baseline 3.

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 precise verb+resource (computes insurance deadlines under the Civil Code) and enumerates the concrete computations: three-month claim-handling limit, 15-day payment due date, termination notice windows, premium-increase objection. It clearly scopes itself against the sibling promlceni by deferring limitation-period computation to it.

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 explains when each mode applies and the contractual-era conditions under which it will and will not compute (Act 37/2004 for 2005–2013 contracts; pre-2005 contracts not handled; pre-2014 dates rejected), and points to promlceni for limitation. No explicit comparison to generic siblings such as lhuta, so it falls short of a full when/when-not/alternatives statement.

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 read-only, idempotent, non-destructive behavior, but the description adds rich behavioral context: it warns above 150 overtime hours, above 300 agreement hours, raises average earnings below the minimum wage, notes that guaranteed wage only triggers a warning for years through 2024, and does not calculate higher work groups.

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?

It front-loads the core purpose but then becomes a single dense paragraph with many legal citations and parentheticals, making it hard for an agent to scan. It is comprehensive, yet not optimally concise or structured.

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

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

Given 15 parameters, a nested object, and no output schema, the description is unusually complete: it covers legal basis, year range, calculation variants, warnings, limitations, and exclusions. While it does not specify a return format, the computed outputs are clearly inferable from the text.

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 contextualizes the calculation but adds little per-parameter meaning beyond what the schema already documents, including the nested sjednane_sazby object and enum values.

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

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

The description opens with a specific verb+resource: 'Zkontroluje mzdu ... proti minimální mzdě a spočítá doplatek' and then enumerates the exact premium categories. It clearly distinguishes the tool from generic wage/earnings siblings by limiting itself to minimum-wage compliance and statutory premiums.

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 explicit conditions: validity years 2017–2027, treatment of pracovni_pomer vs dohoda, monthly vs hourly wage, shorter working time, overtime limits, and premium scenarios. It also states exclusions, e.g. guaranteed wage does not exist in the private sector from 2025 and zaručený plat in the public sector is not calculated.

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

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

Annotations already declare readOnly/idempotent/non-destructive, so the bar is lower; the description still adds real context: only guarantor-checked published pages are returned (drafts excluded), output is Markdown, and long pages are chunked at 30000 characters with continuation via od_znaku. It also discloses that articles include author and date and guides include the guarantor. Auth requirements and failure modes are not covered.

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 action and return format, then content types, exclusions, metadata, and chunking in a logical order. Slightly long with a bilingual EN echo, but no sentence is wasted and the duplication is minimal.

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 and only two parameters, the description carries the return-value burden and does so: format (Markdown), content-type coverage, metadata fields, draft exclusion, and pagination behavior. An agent has everything needed to call and consume the result 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 baseline is 3, but the description meaningfully contextualizes od_znaku by explaining the 30000-character chunking that motivates it and confirms url accepts a full address or just a path. This adds operational meaning beyond the schema's own wording.

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 ('Vrátí celý text publikované stránky ... jako Markdown') and enumerates the page kinds it covers (guide, blog article, service page). It also names the sibling tool hledat_v_pruvodcich as the way to obtain addresses, so an agent can distinguish it from siblings without opening schemas.

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 use: fetch the text of a published page, with addresses sourced via hledat_v_pruvodcich. It routes to the alternative for URL discovery, but does not state an explicit 'when not to use' or a prerequisite that the page must already exist and be published beyond the drafts exclusion.

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

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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive), and the description goes well beyond them: no waiting days, 60% rate, compensation not due if incapacity ended the same day, the 2027 edge case that returns a reason instead of a figure, and the minimum-wage floor. These are real behavioral traits an agent cannot infer from 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?

The core purpose is front-loaded, which is good, but the body is a single very dense block packed with statutory citations and repeated threshold detail. For a complex legal calculator some depth is justified, yet the 2027 caveat and reduction-threshold explanation are verbose enough that trimming would help.

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 comes back, and it largely does: the 60% of reduced earnings, the working-day basis, the 'no compensation due' outcome, and the error-reason path for 2027. It stops short of describing the response shape, but the substantive return behavior is covered.

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 bilingual per-parameter docs, so the schema already carries the load. The description adds interpretive context around inputs (working-day count cannot exceed calendar days, earnings floor, year-dependent reduction thresholds) but little that is not already in the schema, so 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?

States a specific verb and resource with unusual precision: calculates the Czech employer-paid wage compensation for the first 14 calendar days of sick leave/quarantine under §192. An agent can distinguish this from the sibling nemocenska (insurance benefits) by the described scope, but the description never names or contrasts a sibling explicitly, so it stops short of a 5.

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 scope conditions are given: it computes for incapacity arising from 1 July 2019 onward, and for earlier cases it only reports the then-applicable rules without calculating. This tells the agent when the tool produces a real result vs. a historical note. It never names an alternative tool or states when to prefer a sibling, so no explicit 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 this is a safe, read-only, idempotent, closed-world computation, so the safety profile is covered. The description adds genuinely useful behavior beyond that: it estimates statutory rather than contractual fees, enumerates the applicable tariff windows, defines the flat 300 CZK self-representation compensation, and states that output separates what you pay the opponent from your total loss. It does not detail precision or rounding, which is a minor gap.

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 purpose is front-loaded in the first sentence and the dense paragraph is justified by an 11-parameter legal calculator with several version/edge cases. It borders on long and could be broken into tighter clauses, but each sentence carries substantive facts and there is no filler.

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

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

For a complex, configurable calculator with no output schema, the description does the necessary work: it defines scope, the statutory-basis caveat, the applicable date windows, and what the result conveys. The main omission is that it never states a fallback or exclusion route to the narrower sibling tools, but overall an agent has enough to invoke 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 description coverage is 100%, so the schema already documents all 11 parameters in both Czech and English, and the description largely restates the same content (e.g. the advokat_zalobce=false flat-fee behavior also appears in the schema). It adds some cross-parameter context in the narrative but little vocabulary or constraints the schema does not already carry, 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 names a specific verb and resource — it estimates the total cost of a Czech money-claim dispute (court fee, statutory lawyer fee, and who owes/receives what) — and clearly positions itself against siblings like odmena_advokata and soudni_poplatky by stating it computes the statutory tariff fee, not the actual contractual fee. An agent can tell this is the combined dispute-cost calculator rather than a single-component 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?

It gives strong context for when the tool applies: with or without a filing date, the tariff versions that apply from 1 Jan 2014 to 31 Dec 2027, self-represented parties from 6 Oct 2015, and that it does not compute for pre-2014 or post-2027 filings. However, it never explicitly routes the agent to the narrower siblings (e.g. use odmena_advokata for just the fee, soudni_poplatky for just the fee); the differentiation is implied rather than stated as an 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 declare a safe read-only, idempotent, non-destructive profile, so the bar is lower. The description adds real behavioral context beyond that: it returns a reason instead of a result for 2027 cases above the first threshold, returns historical terms without a calculation for pre-2018 cases, and caps the sum at 380 days.

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

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

Front-loaded with the core purpose and rates, and every legal detail (thresholds, rate changes, 2027 caveat, excluded low-income employees) is task-relevant. It is nonetheless a very dense single paragraph that repeats some year/threshold data already present in the schema, slightly hurting scannability.

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

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

For an 8-parameter calculator with no output schema, the description supplies the full legal context an agent needs: formula components, rate schedule, threshold versioning, edge-case fallbacks, the alternative tool, and the 380-day cap. Nothing material for correct 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 coverage is 100 %, so per the rules the baseline is 3. The description earns above baseline by explaining that the year of onset determines the applicable reduction thresholds and rates, and by tying the calculation to the 12-month assessment base and reduction bands, which enriches interpretation of rok and the income parameters.

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 in Czech and EN: it calculates Czech sickness benefit (nemocenská) from day 15 of incapacity, including rates and reduction thresholds. It explicitly distinguishes itself from the sibling nahrada_mzdy_pri_nemoci, which handles the first 14 days, so an agent can route correctly 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?

Gives explicit scope (day 15 onward, employee or OSVČ), valid years (2018–2026 fully, 2027 conditionally), and names the alternative tool for the first 14 days. It also states when-not: employees with gross wage below the reference income are not computed, and earlier disabilities get historical rules without calculation.

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, idempotentHint, and non-destructive behavior, so safety is already covered. The description adds substantial domain behavior: date-dependent tariff versions (from 1 July 1996, 2025 vyhláška), DPH rules, flat-rate fees, and explicit exclusions.

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 calculation purpose and key components before historical details. The Czech paragraph is dense but each sentence carries legal or behavioral detail, and the English summary is efficient. Slightly verbose, but justified for a complex tariff calculator.

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?

No output schema exists, but the description fully explains what is computed and what is excluded, including historical versions and DPH treatment. The return format is not described, but for a pure calculator this is a minor omission.

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 parameter (tarifni_hodnota, plne_ukony, dph, datum_ukonu, polovicni_ukony) is already thoroughly documented. The description only restates high-level inputs and does not add syntax or semantics beyond what the schema provides.

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?

Starts with a specific verb 'Spočítá' and resource 'mimosmluvní odměnu advokáta podle advokátního tarifu', and explicitly names sibling naklady_sporu as the same core but with an arbitrary number of acts. An agent can tell it apart from alternatives without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

States when to use it (given tariff value and number of acts) and when not to use it (does not determine tariff value, does not calculate multiple persons or §12 increases/decreases). It also names the sibling alternative naklady_sporu explicitly.

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 readOnly/idempotent/non-destructive, so the safety profile is covered. The description adds genuine context beyond that: the decree-version selection rule driven by zahajeni_rizeni, the version-dependent behavior (750 Kč/150 Kč pre-1.10.2024, no cap from advances), and the negative scope for liquidator fees. It does not describe the return shape, which is the remaining gap.

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 verb, resource and legal basis in the first clause, then branches by proceeding type. It is a single very dense sentence, but nearly every clause carries regulatory information an agent needs to pick the right enum and inputs, so little is waste.

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 12-parameter calculator with annotations covering the safety profile and a fully documented schema, the description is nearly complete: it maps each proceeding type to its fee rule, explains the date-driven version switch and the out-of-scope cases. The only omission is what the tool actually returns (fee amount / breakdown), which no output schema supplies.

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 every one of the 12 parameters carries a detailed Czech/English description with the governing paragraph. The description restates several of these rules (per-asset secured proceeds, 45 000 Kč minimum, VAT under § 38) but adds no syntax or format detail beyond the schema, so the baseline of 3 applies.

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

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

States a specific verb and resource ("Spočítá odměnu insolvenčního správce") and pins the legal basis (Decree 313/2007, version from 1 Oct 2024). It further enumerates every proceeding type it covers and explicitly declares an out-of-scope case ("Odměnu likvidátora nepočítá"), so an agent can place it precisely among the calculator 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?

Gives clear applicability context: it computes for proceedings begun 1 Jan 2014 – 30 Sep 2024 under the then-current text, and explicitly does NOT compute before 1 Jan 2014 or after 31 Dec 2027. It does not name an alternative tool, but no sibling calculator overlaps this domain, so there is nothing to route to.

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

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

With readOnlyHint/idempotentHint already covering safety, the description still adds substantial behavior: the exact return values ('splněno, nesplněno nebo nutno posoudit'), the cases that force manual assessment, the reporting deadline under § 38v and the § 38v odst. 3 exception. This is rich context well 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?

Content is dense with no filler, but it is delivered as one enormous run-on sentence in Czech with the English summary appended, making it hard to parse. It is front-loaded with the purpose, yet the structure does not help the agent skim the key conditions.

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

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

For an 8-parameter, read-only tool with no output schema, the description covers the domain rules, edge cases (transitional provisions, business property, threshold cap interactions) and the reported outcome, leaving no obvious gap an agent needs 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 baseline is 3, but the description adds interpretive value: it maps each predmet scenario to its holding period, explains the uhrn 100 000 Kč exemption, the 40 mil. Kč cap and its year-dependent scope, and the zůstavitel inheritance rule, all of which clarify how the parameters combine.

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: assesses the holding-period test for exemption of an individual's sale income under § 4 of the income tax act. It clearly distinguishes itself from siblings like danove_odpisy or dar_a_dedictvi_dan and explicitly bounds scope (sales 2021–2027, 'Daň nepočítá').

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 applicability window (prodeje od 1. 1. 2021 do 31. 12. 2027, krypto only from 15. 2. 2025) and the exclusion of tax computation give strong context for when this tool applies. However it never names an alternative tool for the excluded tax calculation, so routing is only implied.

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 safety (readOnly, idempotent, non-destructive, closed world), and the description adds substantial behavioral detail beyond that: it returns both calculation orders because the law is ambiguous, handles the limit case by not applying underinsurance, and specifies date-dependent behavior (2005–2013 same result with citations, pre-2005 no result, pre-2024 warning). This is exactly the kind of legal-logic transparency an agent needs.

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 main purpose and then adds necessary legal nuance. It is somewhat long and dense as a single paragraph, but each sentence conveys specific legal/behavioral information rather than filler, so it is appropriately sized for a complex legal calculator.

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 complexity (multiple legal acts, date-dependent rules), rich schema, and annotations, the description is nearly complete: it explains calculation logic, edge cases, and legal citations. The main gap is that it does not describe the return format or output structure, which is more important since no output schema exists, but it does mention that both calculation orders are returned.

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 all 8 parameters thoroughly. The description adds little parameter-specific meaning beyond what is in the schema; its extra detail is mostly behavioral (calculation order, date handling) rather than clarifying individual parameter semantics. 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 ('Spočítá') and resource ('pojistné plnění z pojištění majetku při podpojištění a po odečtení spoluúčasti'), with precise legal scope. It clearly distinguishes this property-insurance underinsurance calculator from other legal/financial tools in the suite by topic and function.

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 provides clear context for when to use the tool: for calculating property insurance indemnity with underinsurance and deductible, including how optional contract date selects different legal regimes. It does not explicitly name alternative tools or state when not to use this one, but the scenario is well defined.

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 pure read-only, idempotent, closed-world calculation. The description adds substantial behavior beyond that: age-band thresholds (5/8/11 months, the 52/57 edge cases), percentage tiers, rounding-to-crowns-upward, the 0.8× average-wage cap, the fixed-rate fallback when only substitute insurance periods exist, and the input-provenance rule that the tool never derives gross from net.

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

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

Purpose and the 2026 rules are front-loaded and the English one-liner closes it cleanly, but the body is a dense statutory wall enumerating percentages, sections and thresholds that an agent does not need in order to select or invoke the tool. Each sentence is factually relevant, yet several are better placed in output or documentation than in the 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 version-switched statutory calculator this covers the versioning, the age-driven benefit duration, the fallback paths and the input constraints, which is everything needed to invoke it correctly. Although no output schema exists, the description still conveys the shape of the result (monthly amounts, cap, fixed-rate case).

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 meaning beyond the schema by tying 'vek' to the day-of-application age bands, tying 'cisty_vydelek' to the employer certificate and explicitly ruling out gross-pay estimation, and explaining the 'skoncil_sam' / 'vazny_duvod' interaction that only applies to 2011–2025 filings.

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 — 'Spočítá podporu v nezaměstnanosti podle zákona o zaměstnanosti' — and immediately scopes it to statutory versions (2026 rules and 2009–2025 rules). An agent can tell this apart from sibling calculators (prumerny_vydelek, nemocenska, srazky_ze_mzdy) without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

Clearly frames the operating context: verified values for 2026 applications, computation under earlier wording for 2009–2025, and the requirement that net earnings come from the employer certificate rather than an estimate from gross pay. It does not name a specific alternative tool or state when not to use this one, so 4 rather than 5.

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

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

Annotations already declare readOnly, idempotent, non-destructive and closed-world, so the safety profile is covered. The description adds non-obvious behavior: fuzziness/tolerance of typos, refusal to substitute an unrelated entry, and the offered partial/similar-term fallback. It stops short of describing pagination or response size, but for a read-only glossary lookup that is minor.

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 what is returned and what inputs are accepted, with no filler. The bilingual CZ+EN pairing duplicates content, which is defensible for audience coverage but does cost some tightness.

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

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

With no output schema, the description carries the burden of describing returns and it does so concretely (short definition, verbatim statutory text with version, related terms, guide/calculator pages). Behavior on no-match is also covered, leaving nothing an agent needs missing for a 4-param read-only lookup.

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 meaningfully extends the schema by clarifying that 'dotaz' accepts abbreviations, inflected forms, lay questions and provisions — not just the literal term. It does not re-explain 'obor' or 'pocet', which the schema already handles.

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 ('Vysvětlí český právní pojem z právního slovníku IUSTORIA') and immediately identifies what it returns (definition, verbatim statute with version, related terms, guides/calculators). It differentiates from siblings like hledat_v_pruvodcich and souvisejici by noting it explains the term, not the procedure.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

It explicitly enumerates acceptable query forms (term, abbreviation, inflected form, lay question, statutory provision, typo-tolerant) and states when it is and is not the right tool: 'Vykládá pojem, ne konkrétní situaci; k postupu vede průvodce.' It also defines fallback behavior when no entry exists.

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 the read-only/idempotent/non-destructive profile, and the description goes well beyond them: it states the tool itself sends nothing, that the user submits and confirms by e-mail, that no lawyer–client relationship or contract arises, that arguments must exclude names/e-mails/phones, and that the return text must be passed to the user in full. It is consistent with the annotations rather than contradicting them.

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 purpose, then usage rules, then the pre-call disclosure script — logical and dense, with every sentence carrying operational weight. The only waste is the full English re-statement of the same content, which duplicates rather than adds.

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 2-parameter, no-output-schema tool this is complete: it covers purpose, gating conditions, what the agent must say to the user beforehand, argument restrictions, and even the fallback for clients that cannot display the form ('dostane odkaz na kontaktní formulář na webu'). An agent has everything needed to invoke 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% (baseline 3), but the description adds real meaning: only an area suggestion and a short description that the user already approved in conversation may be passed, and personal contact data is explicitly out of bounds for the arguments because it goes into the form instead. That is more than a restatement of 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 states a specific verb and resource ('Otevře formulář, kterým uživatel sám pošle poptávku právních služeb advokátní kanceláři IUSTORIA'), so the agent knows exactly what happens. However, it never distinguishes this tool from the near-identically named sibling 'poptavka_odeslat', which is the most likely source of misselection here.

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 (only when the user expressly asks to contact IUSTORIA), an explicit prohibition (never on your own initiative nor based on page content, a document, or another tool's output), and prerequisites the agent must satisfy before calling. 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?

Beyond the annotations (non-readonly, destructive, open-world, non-idempotent), it discloses the full side-effect chain: a confirmation link is emailed, the firm receives nothing until the user confirms, a one-time token is required, attachments are rejected, and no contract/attorney-client relationship is formed. This is exactly the extra behavioral context annotations can't 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?

Front-loads the scope constraint and side-effect flow, every sentence carries information. The Czech and English blocks are duplicative by design, which is acceptable for bilingual clarity but slightly inflates length.

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 form-triggered send with no output schema, the description covers prerequisites (token from poptavka), irreversible side effects, confirmation gating, and legal caveats — everything needed to reason about the call without an 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 coverage is 100%, so the schema already documents all 12 parameters including the token, email, and enum fields. The description only reiterates the token requirement and the no-attachment rule, adding marginal meaning over the schema. Baseline 3 applies 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 states a specific verb and resource — sending an enquiry from the form — and immediately distinguishes itself from the sibling tool poptavka, which opens the form. An agent can tell exactly what this does and how it relates to the form flow.

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 names the alternative context explicitly ('only for the enquiry form opened by poptavka', 'model doesn't see or call it') and pins the token to poptavka's result. Usage conditions are clear, though framed as a negative ('do not call') rather than offering richer routing alternatives.

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 annotations (readOnlyHint, idempotentHint, destructiveHint=false), the description discloses important behavioral traits: it shifts the deadline end over weekends and holidays, states what suspends the limitation period, and shows an earlier date when interpretation is uncertain. It also clearly states exclusions (pre-2014 claims) and non-computation of amount.

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 then adds behavioral rules and exclusions in a logical order. It is somewhat verbose and partly repeats claim types already listed in the schema's enum, but every sentence contributes useful context for a complex legal 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 complex tool with 17 parameters and no output schema, the description covers purpose, behavioral rules, exclusions, and an alternative tool for debt amount. However, the return format is only vaguely hinted at ('ukáže dřívější datum', 'řekne, co lhůtu zastaví'), leaving the agent without a precise picture of the output 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%, so the schema already documents all 17 parameters in detail, including enum values and their meanings. The description adds no new parameter syntax or format details beyond what the schema provides, so the baseline of 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 ('Spočítá') and resource ('kdy se podle občanského zákoníku promlčí nárok'), then enumerates the claim types it covers. It explicitly distinguishes itself from the sibling tool 'uroky' for debt amount and excludes pre-2014 claims, so an agent can differentiate it from siblings without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

Provides explicit when-not conditions: it does not determine how much the debtor owes (deferring to 'uroky') and does not calculate claims or periods before 1 January 2014. It also explains that it adjusts deadlines for weekends/holidays and what suspends the period, giving clear context for use.

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, idempotent operation, so the description wisely focuses on domain behavior: it explains the reference quarter, the minimum-wage years covered (2017–2027), the special handling for 38.75/37.5-hour regimes, and the 2020–2024 warning about guaranteed wages not being calculated. It also discloses the 21-day cutoff that suppresses a result.

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 and then layers legal qualifications, which is appropriate for a complex statutory calculator. The Czech portion is a long, dense sentence, but every clause carries substantive legal or behavioral information, and the English sentence provides a clean summary.

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 seven parameters, full schema descriptions, and no output schema, the description does the important work of explaining what the tool returns at a high level (hourly average, monthly average, minimum-wage adjustment, holiday pay) and when it will not return a result. It does not describe the exact output object structure, but it is otherwise complete for 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%, so the baseline is 3, but the description adds meaning beyond the schema by giving the legal formula behind the monthly average (weekly hours × 4.348) and the § 357 minimum-wage uplift rules. It reinforces which gross-wage components enter the calculation and how the reference period works.

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 calculation (Czech average earnings) and enumerates the exact outputs: hourly average, minimum-wage adjustment, monthly average, and holiday pay. It also states its role as input for dovolená, náhrada mzdy při nemoci, příplatky, and odstupné, which distinguishes it from those sibling 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?

The description clearly explains its role as the upstream calculation for several sibling tools and notes the key exclusion that fewer than 21 worked days yields no result (pravděpodobný výdělek). It does not explicitly name an alternative tool to use for probable earnings or for the downstream calculations, but the context is sufficient for 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 readOnly/idempotent/non-destructive, so the safety profile is covered. The description adds real behavioral context beyond that: date-dependent tariff-version selection (2026 ruleset vs. 2014–2025 ruleset), the special-case rules for divorce/appellate review dates, and the fact that it also reports refunds on settlement or withdrawal.

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 purpose, then the coverage list, then the date/version rules — nearly every sentence encodes a concrete rule. It is somewhat long, partly because the English line restates the Czech, but the density is justified.

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

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

For an 8-parameter, no-required-parameter calculator with no output schema, the description covers purpose, scope limits and version selection well, and hints at the return value ('kolik se vrátí'). It stops short of describing the returned amount/format, which is the only notable 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 the schema already documents all 8 parameters, their enums and defaults. The description reinforces key semantics (the filing date selects the tariff version), but adds little syntax or meaning the schema does not already carry; 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 precise verb ('Spočítá' / calculates) plus the exact resource (court fees under the court fees act) and enumerates the covered subject matters: money claims, divorce, property/co-ownership settlement, ownership determination, appeal and appellate review. An agent can immediately tell this apart from attorney-fee or litigation-cost siblings.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

The scope enumeration effectively communicates when to use it, and it adds genuine exclusions ('odvolání ve věci rozvodu nepočítá', no computation for proceedings before 2014 or after 31 Dec 2027). However, it never names or routes to a sibling tool, so the alternative-selection guidance is 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?

Annotations already declare readOnly, idempotent, non-destructive, and closed-world. The description adds real value beyond that: results are ordered in a fixed, explainable way with no ML model, every result carries a stated reason, and service pages, price lists, and contact pages are never returned.

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 action and followed by the reason taxonomy, exclusions, and the input constraint. Dense but every clause earns its place; the Czech/EN duplication is the only mild redundancy.

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

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

With no output schema, the description carries the return-value burden and does so by describing results as pages plus a reason, and by stating what is excluded. Input constraints and ordering behavior are covered, so an agent has enough to call it correctly, though sibling differentiation is thin.

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 does map the four input kinds (url, ustanoveni, nastroj, pojem) to the search behavior, which slightly clarifies their semantics, but it adds no format, syntax, or edge-case 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+resource: finds related IUSTORIA guide pages and articles for a given page, provision, calculator, or glossary term. It goes further by saying each result comes with a reason and lists the reason types (shared provisions, same guide, direct link, topic), so an agent can distinguish it from sibling search tools like hledat_v_pruvodcich.

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 one concrete usage constraint — 'Zadejte právě jedno z url, ustanoveni, nastroj, pojem' (specify exactly one input) — which is genuinely useful. However, it never states when to prefer this tool over siblings such as hledat_v_pruvodcich, pojem, or zneni_predpisu, so routing remains partly inferential.

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 readOnly/destructive/idempotent annotations, it discloses rich behavior: which legal rules are applied per mode, year-dependent thresholds, refusal of the manzele=true + manzel_s_duchodem combination, and a no-result-with-reason policy for unverified years. This is substantial 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?

The purpose is front-loaded, but the body is a dense, sprawling block of legal prose with many nested clauses that is hard to scan, and parts of it restate parameter semantics already in the schema. The length is partly justified by tool complexity, but structure is weak.

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 13-parameter legal calculator with no output schema, the description covers the computation model, both modes, year handling, spousal/joint-debt-relief rules, and edge-case refusals, so an agent has enough to call it correctly without a return-value spec.

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 per-parameter descriptions are already very detailed, so the baseline is 3. The description reinforces concepts (vyživované osoby, oprávnění z výživného, přednostní, čtyři exekuce, year-based rules) but adds no syntax or format meaning beyond what the schema already provides.

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 (Spočítá) and resource (srážky z čisté mzdy při exekuci / měsíční splátka v oddlužení) with precise legal grounding, making it clearly distinguishable from siblings like exekutorsky_tarif, odmena_insolvencniho_spravce and minimalni_mzda_a_priplatky.

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 clearly lays out the two modes (exekuce, oddluzeni) and the conditions under which each applies, and states a boundary: for years without verified values (2027, pre-2023) it returns no result and explains why. It does not explicitly name sibling alternatives, but the domain context is unambiguous.

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 destructiveHint=false, so safety is covered. The description adds substantial behavioral context beyond that: which legal regime applies by period, that pre-2016 periods are unsupported, and which secondary outputs (audit, deadline, retention) are produced.

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 purpose is front-loaded and every clause references a concrete legal rule, but the Czech body is extremely dense with nested conditions, and the appended English recap partially duplicates it. Structure is serviceable rather than clean, and an agent must parse heavy legal prose.

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 statutory calculator with 8 parameters, a nested object and no output schema, the description covers the computation rules, the applicable law by period, exclusions and the set of derived results. It is close to complete, though the precise output shape must be inferred.

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 all 8 parameters, giving a baseline of 3. The description adds some interpretive context (turnover recalculated by month count, previous-period values needed for the old § 20 audit test) but does not restate parameter formats or units beyond what the schema provides.

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?

Specifies a precise verb+resource: determines the Czech accounting entity category, audit duty, filing deadline, penalties and retention period. It is clearly distinguishable from siblings like lhuta, danove_sankce, or danovy_kalendar, and even names the legal basis (§ 1b) that defines the computation.

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 the applicable scope (obchodní společnosti a družstva) and an explicit exclusion ('Období započatá před rokem 2016 nepočítá'), and routes periods to the correct legal version. It does not name a sibling alternative, but the domain is narrow enough that implied usage is unambiguous.

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 the safety profile (readOnly, idempotent, non-destructive), so the description must carry behavioral disclosure – and it does: it warns it will NOT substitute an unknown ČNB rate (result then contains no number), refuses to compute interest on interest, flags § 1805/§ 580/§ 588 signals without concluding invalidity, and applies the § 1932/1933 payment ordering. These are material behaviors 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?

The core purpose is front-loaded, but the body is an extremely dense single run-on sentence packed with statutory citations and parentheticals, which makes it hard to scan. Given the domain complexity some length is warranted, but it is far from tight.

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, output-schema-less legal calculator the description is remarkably complete: it explains the regimes, the flat fee, the payment-allocation rules, and even what the tool will and won't output. It stops short of describing return structure (no output schema exists), keeping it from a 5.

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 itself documents every field including enums and date patterns, so the description's parameter discussion largely restates legal background rather than adding invocation syntax. 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 opens with a specific verb+resource ("Spočítá úroky podle českého práva s rozpisem po úsecích") and enumerates the exact interest regimes it handles, so an agent can immediately tell it apart from the other legal-calculator siblings (odmena_advokata, danove_sankce, etc.). The scope is concrete and 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 rich context for when each mode applies (statutory vs agreed default interest, contractual loan interest, maintenance interest, B2B flat fee), effectively telling the agent which regime maps to which situation. However, it never states exclusions or explicitly routes to any alternative tool, so it falls short of a full when/when-not statement.

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/destructiveHint false and openWorldHint false, but the description adds substantive context they don't carry: the document is assembled in the user's browser and filled data never leaves to the IUSTORIA server or the conversation. That privacy/flow detail is exactly the kind of behavioral disclosure annotations cannot express.

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 behavioral and usage guidance is front-loaded in tight sentences before the enumeration. The 29-item template list dominates the length, but each entry carries information the enum alone does not, so it largely earns its place rather than being 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 single-enum, read-only template-opening tool with no output schema, the description covers purpose, when to use it, the rendering fallback, the privacy flow, and the meaning of every valid id. Nothing an agent needs to invoke 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 coverage is 100% and the enum lists slug identifiers only, so the schema cannot tell the agent what any slug means. The description supplies Czech human-readable meanings for all 29 template ids (e.g., 'uznani-dluhu (Uznání dluhu a dohoda o splátkách)'), which is genuine added meaning beyond the structured 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?

States a specific verb and resource: it opens a template form in the conversation for the user to fill in and download as .docx. It also distinguishes itself from vzor_info by delegating 'what it's for and when not to use it' to that sibling. An agent knows exactly what this does without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicit when/when-not/alternatives: do not fill the form on the user's behalf, do not ask for personal data in chat, and consult vzor_info for purpose and non-use cases. The fallback for clients that cannot render the form (a web link to the same form) is also specified, which is unusually complete routing 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 cover the safety profile (readOnly, idempotent, non-destructive, closed-world). The description adds genuinely new behavioral context: the template is NOT filled server-side, it is composed in the user's browser, and entered data never leaves the client. That is useful information 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?

Front-loaded with the essential purpose and the client-side-filling caveat before the long template list. The enum enumeration is long but earns its place by giving descriptions the schema lacks; otherwise there is 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 no output schema, the description correctly explains what the tool returns (template purpose, required content, exclusions, generator link) and clarifies the client-side filling behavior. An agent has enough to call it correctly; only minor details of the response format are unspecified.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% but the schema exposes only raw enum ids. The description maps all 29 ids to human-readable meanings (e.g. 'kvitance (Kvitance)', 'reklamace (Reklamace zboží)'), which materially exceeds what the schema provides.

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 ('Řekne, k čemu vzor dokumentu z průvodců IUSTORIA slouží...') and enumerates exactly what it returns: purpose, required content, when not to use, and a link to the web generator. It implies its informational nature, but never names the closest sibling (vzor_formular), so differentiation is left implicit.

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 'kdy ho nepoužít' refers to when not to use a template, not when to call this tool versus an alternative. Usage is only implied (call it to learn about a given template); no alternative tool such as vzor_formular is named or contrasted.

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

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

Annotations already declare this a safe, idempotent, read-only calculator, and the description goes well beyond by disclosing concrete behavior: the social advance is capped at 1/12 of the 48x-average-wage maximum base and 'řekne to', while health insurance has no cap; it also summarizes the 2026 minimum-base change, refund mechanics, and what it refuses to compute. That is substantial behavioral context not present in 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?

Front-loaded with the core purpose and scoping, followed by capabilities and exclusions; every sentence is informative. It is dense and long, with legislative citations that could be trimmed, but nothing is purely filler.

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

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

For a complex, 11-parameter, multi-mode calculator with no output schema, the description is complete: it covers scope, supported years, mode semantics, edge cases (new OSVČ, exemptions, caps), and honest limitations. An agent has everything needed 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 description coverage is 100%, and the 11 parameters are each richly documented in the schema itself (enums for rezim/cinnost, defaults, ranges). The description reinforces the domain but adds little parameter-specific semantics beyond what the schema already provides, so 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?

States a specific verb and resource: calculates minimum and actual monthly OSVČ advances for social (29.2%) and health (13.5%) insurance, or the annual contribution from the tax base. It even names the exact legal rates and modes, so an agent knows precisely what domain this covers and can distinguish it from siblings like minimalni_mzda_a_priplatky or nemocenska.

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 each regime (monthly advances vs annual pojistné), the years covered, and explicit exclusions ('Nepočítá paušální režim, daňový základ ani souběh se zaměstnáním u maximálního základu'). It does not, however, name a sibling tool to use for the excluded cases, so it stops short of full when/when-not/alternative 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 cover readOnly, idempotent, non-destructive, and openWorld=false. The description adds that it explains version binding without calculation, that the server interprets law from 1 Jan 1993, and that it returns transitional provisions and sources. It does not contradict annotations. It adds meaningful behavioral context 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?

The description is dense but front-loads the core purpose in the first sentence, then adds nuance about calculators, the time bound, and parameter modes. Some redundancy between Czech and English sections, but that's for bilingual accessibility. Efficient overall, though could be slightly tighter.

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, a 3-param tool with 0 required and 100% schema coverage, the description provides essential context: what the tool returns (version, concept, transitional provisions, sources), the temporal limitation (since 1993), and the two usage modes. It is nearly complete; minor gap in not describing output format, but it lists components. Adequate.

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 schema already documents parameters thoroughly, including enums with extensive mappings in the predpis description. The description adds high-level guidance: 'Zadejte nastroj ... nebo predpis ...', indicating the two modes. It doesn't add syntax or format beyond the schema. Baseline 3 is appropriate when schema is this rich.

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 ('Řekne, které znění...') and resource ('znění předpisu' for a calculator) and explicitly notes it does not calculate, distinguishing it from the many calculation siblings. It is clear. However, it doesn't explicitly name a sibling alternative (e.g., 'if you want the result, use the calculator') though it implies it. The purpose is well-served.

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 (to know which version applied on a day without calculating) and when not to (it does not calculate; calculators do). Provides the alternative implicitly: use calculators for results. Also gives parameter usage guidance: enter either nastroj or predpis. This is strong usage guidance.

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