Descontinho

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

Annotations already declare readOnly/idempotent/non-destructive, and the description adds genuinely non-obvious behavior: a 7-day history window, a time-weighted-median detection method, and the rule that the seller's advertised 'de' price carries no evidential weight. It does not cover pagination, ordering of results, or volume/cost limits, which are meaningful for a list tool capped at 50 items.

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 the certainty semantics are front-loaded, which is right, but the same point is made twice – 'o "de" do vendedor não conta' early and 'O preço "de" que o vendedor escreve... não sustenta desconto nenhum aqui' later – and the closing warning about citing percentages is a third pass at the same idea. Tightening would gain real space without losing meaning.

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 usefully documents the one field that governs downstream use, discount.certainty, plus the freshness of the underlying data. It stops short of describing the overall result shape or how the 6 optional parameters interact (e.g. empty-result behavior when busca finds nothing), but for a read-only filtered list this is close to sufficient.

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

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

Schema coverage is 83%, so the schema already documents busca, categoria, preco_maximo, desconto_minimo and apenas_verificados, and the baseline of 3 applies. The description complements desconto_minimo conceptually ('queda real contra o preço típico que medimos') but adds no syntax, format or interaction detail beyond the schema, and says nothing about limite or the deprecated apenas_verificados flag.

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 – lists Shopee Brasil products whose price drop is proven by the tool's own 7-day reading history – and explains the verification mechanism, which is more than a restatement of the name. However, it never distinguishes itself from the sibling quedas_de_preco, which by name covers the same territory; an agent cannot tell which one to open without inspecting the 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?

Strong guidance exists on how to consume results – only 'verified' certainty may be presented as savings, and quoting a percentage without the certainty label misinforms the reader. But that is output-interpretation guidance, not tool-selection guidance: there is no statement of when to use this versus historico_de_preco or quedas_de_preco, which is the gap that matters most for an agent choosing a tool.

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 read-only, idempotent, closed-world profile, so the bar is lower. The description still adds real behavior beyond them: it accepts either an itemId or a full URL, it returns a reading count alongside the series, and the response explicitly flags the under-three-readings case as insufficient evidence for a decline — a non-obvious output contract.

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?

Three short sentences, front-loaded with what the tool returns, then the input forms, then the caveat. Nothing is repeated and nothing is 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 one-parameter read-only tool with no output schema, the description covers input forms, the returned series, and the reading-count caveat. The remaining gap is interpretive: it never states the period or granularity the series spans, which matters for judging a price trend.

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 single parameter already documents both accepted forms (numeric itemId or full product URL) with an example format. The description restates exactly that, adding no syntax, validation, or edge-case detail beyond the schema, so 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 concrete resource — the observed price series for a Shopee product — plus the accepted identifier forms and the extra reading-count field. It never names quedas_de_preco or buscar_promocoes, so the boundary between 'give me the raw series' and 'give me detected drops' must be inferred from the wording.

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

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

Usage is only implied: the note that fewer than three readings cannot support a drop claim hints at when this raw-series tool is the right primitive, but the description never says when to call it instead of the sibling quedas_de_preco (which presumably already applies that rule). No exclusions or prerequisites are stated.

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

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

Annotations already cover safety (readOnly, idempotent, closed-world), yet the description adds substantial behavior of its own: the detection methodology (time-weighted median so flash promos only count the hours they lasted), the 7-day tracking floor that explains sparse results, and the meaning of the returned discount.certainty flag. The warning that quoting the percentage without certainty misinforms is genuine usage-critical context.

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 qualifying rule is front-loaded in the first sentence, and the certainty explanation follows where it matters most. Slightly verbose, with some redundancy between the certainty definition and the closing warning about citing percentages, but no sentence is wasted.

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 available, the description carries the burden of explaining return values and does so by defining discount.certainty as 'verified' vs 'none' — the one field an agent could easily misuse. Combined with the 7-day prerequisite, an agent has what it needs to call this and interpret the results responsibly.

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?

There is exactly one parameter (limite) and the description never mentions it, despite 0% schema description coverage. The schema's default of 10 and bounds of 1-50 make the meaning self-evident, so this is a minor gap rather than a blocking one — baseline territory.

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 the resource precisely (products) and the exact qualifying rule: a drop of at least 15% and R$ 20 versus the time-weighted 14-day median, or an all-time low. An agent knows exactly what comes back. It does not name the siblings (buscar_promocoes, historico_de_preco) or explicitly contrast itself with them, 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?

It defines the qualifying conditions well (thresholds plus a 7-day tracking prerequisite), which implicitly tells the agent when results will be empty. However there is no explicit "use this instead of X when..." routing to buscar_promocoes or historico_de_preco, and no stated exclusions, so usage is only implied.

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