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

Annotations already declare readOnlyHint=true and openWorldHint=false, so safety is covered. The description adds genuinely useful context beyond that: the output is model interpretation (Gemini 3.5 Flash-Lite), tone and factors are subjective and should be cross-checked. It does not mention latency or cost of the AI call, which would have made it 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?

Two sentences, front-loaded with what the analysis contains before the verification guidance. Dense but every clause carries information; the model name is arguably dispensable but minor.

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 output schema exists so return values need no explanation, and the description still summarizes the analysis contents. Trust caveats and sibling routing are covered, leaving nothing an agent needs in order 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% and the single parameter 'reuniao' is documented in the schema as the meeting number. The description adds no additional meaning about the parameter (e.g., valid range or where to obtain the number), 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+resource: an AI-generated analysis of a meeting minute, enumerating exactly what it returns (decisão, tom, fatores, riscos, orientação futura, parágrafos citados). This distinguishes it clearly from siblings like ler_ata (raw text) and listar_reunioes (official numbers).

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 routes the agent: use ler_ata to verify the interpretation against the source minute, and use listar_reunioes for official numbers (Selic before/after, change). It states both the tool's role and the alternatives for adjacent needs.

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

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

Annotations already declare readOnlyHint=true and openWorldHint=false, and the description adds value beyond that by disclosing the shape of returned excerpts and the query-retry behavior for technical vocabulary. It does not discuss result limits, ranking, or empty-result behavior beyond the retry hint.

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?

Two short paragraphs, front-loaded with what is returned, followed by actionable retrieval advice. Every sentence carries weight, including the concrete vocabulary substitution examples.

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 an output schema covering return shape and annotations covering the safety profile, the description fills the remaining gaps: query strategy and excerpt usability for citation. It leaves the paging/limite behavior to the schema, which is acceptable but slightly thin for a search tool.

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

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

Schema description coverage is 80%, so desde/ate formats, reuniao restriction and the consulta example are already documented. The description adds usage meaning for desde/ate ('comparar períodos') but says nothing about limite or how parameters combine, so the schema still carries most of the 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?

States a specific verb and resource ('Busca parágrafos das atas do Copom') plus the scope ('sobre um tema') and the returned unit ('trechos com reunião, data, seção e parágrafo para citar'). The paragraph-level theme search is naturally distinct from siblings like ler_ata (full meeting) and listar_reunioes (meeting listing), so an agent can route 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?

Gives concrete operational guidance: combine with desde/ate to compare periods, and if the first search misses, retry with Copom-specific vocabulary or split into several short searches. No sibling is named as an alternative or exclusion, which keeps it from 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 readOnlyHint and openWorldHint, so safety is covered. The description adds genuinely useful behavior beyond the schema: long daily series (>400 points) are automatically downsampled to the last value of each month to fit the conversation context — a transformation the agent must know to interpret results correctly.

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?

Two sentences, front-loaded with the resource and source, followed by the downsampling caveat. No wasted words, though the truncation rule could be stated slightly more crisply (e.g., naming the threshold as automatic aggregation).

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 an output schema present, return-value documentation is unnecessary, and annotations cover the safety profile. The description fills the most important remaining gap — the monthly downsampling for long daily series — so an agent can interpret results; only minor gaps remain (frequency defaults, whether values are level or change).

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 67%: inicio and fim are documented in the schema with accepted formats, but the five enum values for indicador carry no explanation in either place. The description reinforces the two-date range concept and "AAAA" granularity indirectly, but does not add syntax or semantic 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 the specific resource and source: "Série oficial do Banco Central (SGS) entre duas datas", which tells an agent it retrieves official BCB time-series data over a date range. The implicit verb (query/retrieve) is clear from the name, and the sibling tools (atas, reuniões) make the domain distinction obvious.

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: it is the sole indicator-query tool among meeting-related siblings, so selection is unambiguous by default. However, there is no explicit statement of when to prefer it, what "indicador" codes mean, or what to expect if the range is invalid — guidance remains implicit.

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

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

The annotations already establish that this is a read-only, closed-world operation. The description adds useful context about the return structure (section/paragraph organization and interest-rate decision), but it does not discuss error behavior, permissions, or retrieval limits.

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 a single, front-loaded sentence with no wasted wording. It efficiently states the resource, the output organization, and the key content included.

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 output schema exists and annotations cover the safety profile, so the description need not explain return values in detail. It adequately states the content and structure of the result, though it would be stronger with guidance on how this differs from buscar_atas.

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 already documented in the schema. The description does not add meaning beyond the schema for 'reuniao' or 'secao', 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?

The description uses a specific verb and resource: it returns the text of a meeting minute, organized by section and paragraph, including the interest-rate decision. This is clear, but it does not explicitly distinguish the tool from the sibling buscar_atas, which likely also deals with minutes.

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?

There is no explicit guidance on when to use this tool versus buscar_atas or analise_reuniao. The required 'reuniao' parameter implies retrieving a specific meeting's minutes, but the description gives no exclusions, alternatives, or prerequisites.

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

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

Annotations already declare readOnlyHint=true and openWorldHint=false, so safety is covered. The description adds genuine behavior beyond that: results are truncated to the `limite` most recent matches while `total` counts all matching meetings, which is exactly the kind of truncation disclosure 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?

Two short paragraphs, zero filler. The returned fields are named first, then the truncation/pagination rule, then the fallback advice — properly front-loaded.

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

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

An output schema exists, so return values needn't be explained, and annotations cover the safety profile. The description supplies the one non-obvious thing (most-recent truncation with `total` for full count); only explicit sibling routing 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 75% and already documents date formats and the enum values for `decisao`. The description still adds meaning: `limite` selects the most recent N (not an arbitrary N) and the date bounds are the correct lever for older data, going beyond the schema's bare 'Máximo de reuniões, as mais recentes'.

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 ('Lista reuniões do Copom') and even names the payload fields returned (meta Selic antes/depois and the decision). It does not explicitly distinguish itself from siblings like buscar_atas or analise_reuniao, so it falls 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 gives real operational guidance — use `desde`/`ate` rather than inflating `limite` for old periods — which tells the agent how to navigate the result window. However, it never says when to use this tool instead of buscar_atas, ler_ata or analise_reuniao, so sibling routing 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.