cenaodhad – ceny bytů v Praze z katastru

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

Annotations already cover the safety profile (readOnly, idempotent, closed-world, non-destructive), so the bar is lower. The description adds genuine value beyond them by disclosing the return semantics (median CZK/m² from realised land-registry sales) and the aggregation behavior (a borough returns its districts, a street returns the districts it runs through). It stops short of noting data recency, coverage limits or what happens for an unknown locality.

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 and segmented with pipe separators into use / do-not-use / scope blocks, which is easy to scan. However, the entire block is duplicated verbatim in Czech and English, roughly doubling length for content that the jazyk parameter already governs.

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

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

For a two-parameter read tool with no output schema, the definition covers purpose, exclusions, aggregation semantics and the return metric. An agent has everything needed to select and invoke it correctly without further context.

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

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

Schema description coverage is 100% and the schema already documents both parameters, including locality formats and diacritics/case tolerance. The description adds no parameter-level meaning beyond what the schema states, 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: returns the median CZK/m² price level for apartments in ONE Prague locality, sourced from realised sales in the Czech land registry. It also names the scope granularity (district, borough or street), which cleanly separates it from the per-apartment and multi-locality 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?

Explicit when-to-use ('user asks about the price level in ONE Prague locality without a specific apartment') plus an exhaustive when-not-to-use list that routes each excluded case to a named sibling (odhad_ceny_bytu, srovnani_ctvrti, top_ctvrti_praha, overeni_ceny_inzeratu) and even splits by property type (domy, pozemky). 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?

Goes well beyond the annotations: discloses the recipient (third-party agent), response SLA (one business day), rate limits (5/hour per IP, shared/higher for Claude+ChatGPT, plus a daily cap), required data, and the failure mode when consent/confirmation are missing (returns a recap and consent text, stores nothing). This is exactly the behavioral context annotations cannot carry.

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

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

Content is dense and front-loaded with the critical caveat ('optional, only at the user's explicit request'), and every sentence carries necessary safety or routing information. It loses a point only because the CZ/EN duplication roughly doubles the length with partially overlapping text.

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, no-output-schema, write-ish tool, the description covers the workflow (summarise details, get confirmation and consent), the gate behavior, the recipient, and rate limits — 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%, so the baseline is 3, but the description adds real meaning: it tells the agent which fields to pass for house/land (typ_nemovitosti, plocha, pozemek_m2 or typ_pozemku, and the mid estimate odhad_kc) and stresses that potvrzeni_odeslani/souhlas_se_zpracovanim must reflect genuine user confirmation rather than being auto-filled.

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 action (passes a consultation request to a named Reality Style agent, Petr Kubačka, Prague) on a specific resource (a property consultation), and explicitly contrasts this with the valuation siblings. An agent can distinguish this from odhad_ceny_* 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?

Explicit when-to-use ('only at the user's explicit request', 'wants a more precise assessment or to be contacted'), when-not-to-use ('Do not use just to get a price or apartment estimate'), and names the alternatives to route to instead (odhad_ceny_bytu, ceny_bytu_lokalita, overeni_ceny_inzeratu). It even specifies the precondition that confirmation and consent must not be filled in on the user's behalf.

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 the description adds substantial extra context beyond them: data source is realised sales from the land registry rather than listings, output is an indicative range and not a certified appraisal, no sign-up or contact details required, and the model deliberately ignores floor level, lift and orientation.

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?

Every sentence carries real content (usage rule, output semantics, exclusions, parameter rule), and the critical constraint is front-loaded in the first sentence. The CZ/EN duplication roughly doubles the length, which is defensible for a bilingual tool but is still redundancy an agent does not strictly need.

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

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

Although there is no output schema, the description itself explains what is returned (mid estimate, range, CZK/m²) and its caveats. For a read-only two-required-parameter tool, nothing an agent needs to call it correctly is missing.

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

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

Schema coverage is 100%, so the baseline is 3, but the description adds genuine meaning: it specifies that ctvrt must be a cadastral area and how to map a street to a district, and it reinforces the area requirement. The 15–250 m² bound and jazyk enum remain schema-only, which keeps this from a 5.

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 (odhad ceny) and resource (konkrétního bytu v Praze) plus the required inputs (čtvrť, plocha v m²), and explicitly distinguishes itself from sibling tools by naming them. An agent can separate this from odhad_ceny_domu, ceny_bytu_lokalita, and srovnani_ctvrti 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 explicit when-to-use, when-NOT-to-use with named alternatives (overeni_ceny_inzeratu for listing prices, ceny_bytu_lokalita for locality-level, srovnani_ctvrti for comparisons), plus a fallback rule when the required area is missing. This is as complete a routing spec as one could ask for.

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

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

Annotations already declare read-only, idempotent, non-destructive, closed-world. The description adds genuinely useful non-annotation context: no contact details required, results are asking prices converted to realised prices anchored to ČSÚ data (not the land registry), and that condition/floor/orientation are ignored. Return format is sketched (mid value + range), though no pagination/precision caveats.

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?

Bilingual duplication doubles length, but each section is front-loaded and purposeful: use-case, inputs, output, then rules and routing. No filler sentences, though the CS/EN mirroring is somewhat 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?

With no output schema, the description carries the return-value burden and does so (indicative mid value and range, source methodology). Prerequisites, exclusions, and sibling routing are all present; nothing an agent needs to invoke 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% with formats, ranges and examples already documented for all four parameters. The description restates the input trio (locality, usable area, plot size) and notes locality accepts čtvrť/ulice/'Praha N' with optional diacritics, but adds little beyond the schema. Baseline 3 is correct.

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 (estimate the price of a family house in Prague) with scope limits and names the siblings it is not (odhad_ceny_bytu, odhad_ceny_pozemku). An agent can distinguish it from siblings without opening a schema.

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

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

Explicit when-to-use, exclusion routing to the two sibling estimators, a rule to ask for missing usable area/plot, and guidance on when konzultace_makler is warranted (only if the user wants it). Also states which attributes not to ask for.

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

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

Annotations already cover readOnly, idempotent, non-destructive and closed-world. The description adds genuinely new behavioral context: values are indicative mid + range derived from asking prices adjusted to realised prices (not registry data), and the call requires no contact details. It doesn't mention latency or failure modes, so it stops short of 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 definition is long but front-loads the trigger case before the routing rules, and every block earns its place (usage, inputs, return semantics, rules). The CZ/EN duplication doubles the text for bilingual coverage, which sacrifices tightness but is justified by the audience.

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 explains what comes back (indicative mid value and range), where the data originates, and what is not required (contact details). Combined with explicit routing and input fallbacks, an agent has everything needed to call it correctly.

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

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

Schema coverage is 100%, so the schema carries most parameter meaning. The description still adds value: it states 'typ' defaults to building plot when omitted, tells the agent to prompt for 'vymera_m2' when absent, and describes acceptable 'lokalita' forms (district, street, 'Praha N').

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 (land price estimate) with concrete scope: Prague land, price per m², building plot focus, and clarified sourcing (asking prices converted to realised prices, not land registry). It distinguishes itself from odhad_ceny_domu and konzultace_makler by name, so an agent can route correctly 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?

Explicit trigger ('when the user asks about price/value of land in Prague') plus explicit routing rules: land with a house → odhad_ceny_domu, exact price → konzultace_makler only if the user wants it, and a behavioral rule to ask for missing plot size or confirm land type. This is a strong when/when-not/alternative 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 readOnly/openWorld=false/idempotent/non-destructive, so safety is covered. The description adds real value beyond them: it lists what is returned (range, middle, % difference, CZK/m², five-band position), frames the output as a neutral comparison rather than a verdict, and discloses a limitation — it does not fetch listings or open links.

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 use case, then the do-not-use routing, then output shape; every sentence carries content. Length is doubled by the CZ/EN mirror, but that duplication is inherent to the bilingual contract and each version is internally waste-free.

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 (range, middle, % format, CZK/m², five bands). Combined with annotations covering safety and schema covering all params, an agent has everything needed to select and 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%, with ranges, examples and the ctvrt/jazyk distinctions already documented in the schema. The description only restates that district, area and price must be supplied by the user, adding no format or constraint detail beyond the structured fields. 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 — comparing an asking price against the range of realised sales of similar apartments in the same Prague district — with the scope (apartment, same district, registry data) made explicit. It is readily distinguishable from odhad_ceny_bytu (no price given) and from the house/land estimators named in the do-not-use clause.

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

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

Explicit when-to-use (buyer has a concrete asking price from a listing) plus when-not (no specific price → odhad_ceny_bytu; houses/land → odhad_ceny_domu / odhad_ceny_pozemku), and it states the precondition that inputs must come from the user or pasted text. Nothing is left to inference.

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

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

Annotations already declare readOnly/idempotent/non-destructive/closed-world, so the safety profile is covered. The description adds substantive behavioral detail beyond that: it returns median CZK/m² from realised land-registry sales, sorted most-expensive-first with % difference vs the top locality, and that streets are unsupported. It stops short of noting coverage limits or data recency.

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 trigger condition and followed immediately by the exclusions, which is exactly the right order. The bilingual duplication roughly doubles length, but each half is dense and waste-free.

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

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

For a two-parameter read tool with full schema coverage and no output schema, the description supplies the return shape (median CZK/m², ordering, % delta) and all routing rules. Nothing an agent needs to call it correctly is missing.

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

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

Schema description coverage is 100%, so both parameters are fully documented in the schema (including the 2–8 array and 'Praha N' format). The description reinforces the district-vs-street constraint but adds no syntax or format detail beyond 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+resource (compare 2–8 Prague localities) with explicit scope limits (2–8, districts/boroughs, streets not supported) and names the sibling tools it is not. An agent can distinguish it from ceny_bytu_lokalita, top_ctvrti_praha and odhad_ceny_bytu 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?

Explicit when-to-use (user wants to COMPARE 2–8 specific localities) and when-not-to-use, each with the correct alternative tool and the selecting condition (single locality, citywide ranking, specific apartment with floor area). Nothing is left to inference.

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

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

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so safety is covered structurally. The description adds genuine context beyond that: the data provenance (median CZK/m² from realised sales in the Czech land registry) and that the result is ordered, plus the discovery use-case for valid district names. It omits any note on result size limits, but the schema's pocet bound largely covers that.

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 bilingual CZ/EN split is justified for the audience and each half is front-loaded with the trigger condition followed by exclusions. There is some duplication of the ranking and 'do not use' content across languages, but no filler within a language.

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 fills that gap by explaining the return shape (districts sorted by median CZK/m²) and its source, and it covers routing, scope and discovery use-cases. The only gap is absence of any note on the 1–20 result-count bound or empty-result behavior, which the schema covers only partially.

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 two enums fully documented, so the baseline is 3. The description only indirectly touches parameters ('one borough' hints at obvod, the ranking framing mirrors kriterium) without adding format, default or boundary 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 action (ranking districts) and resource (Prague districts by median CZK/m² from realised registry sales), with an explicit scope qualifier ('all Prague or one borough'). It clearly separates itself from the sibling comparison and single-locality tools by 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?

Contains an explicit 'Use when' trigger (ranking most/cheapest districts, affordability, discovering valid district names) and an explicit 'Do not use when' clause routing to srovnani_ctvrti for named localities, ceny_bytu_lokalita for one locality, and odhad_ceny_bytu for a specific apartment. Both conditions and alternatives are spelled out.

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