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

Annotations (readOnlyHint, openWorldHint, idempotentHint, destructiveHint=false) already cover the safety profile, so the bar is raised — and the description clears it. It adds the concrete mechanism ('a live GET against the public IBGE SIDRA API'), the return shape ('Markdown plus a typed structuredContent payload'), and several non-obvious behaviors: estatisticas=true ignores pagina/campos/formato, processes ALL rows before pagination, returns an empty registros, excludes SIDRA absence markers, and respects the 100,000-value cap. This is exactly the hard-won context 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?

Well-structured and front-loaded: definition → common tables → territorial levels → examples → complex stats mode → sibling routing → behavior statement. The length is justified by 12 parameters and a genuinely complex special mode, and each section earns its place. It is not maximally tight — the territorial-level list repeats schema content and the statistics paragraph is dense — but the organization compensates.

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 low-level engine with 22 siblings, the description covers everything needed to select and invoke it correctly: purpose, data scope, common table codes, examples, special-mode behavior, parameter interactions, cap limits, sibling routing, and safety profile. An output schema exists, so return-value details need no elaboration here. No material gap remains.

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 and the baseline is 3. The description earns a point above baseline by supplying real table-code examples ('tabela="6579", periodos="2023"', 'localidades="3550308"'), a curated quick-reference of territorial levels, and cross-parameter semantics: estatisticas=true ignoring pagina/campos/formato, agruparPor ranking behavior ('grupos[0] = maior total'), and SIDRA absence markers excluded from n. The territorial-level list partially duplicates the schema's nivel_territorial description, which prevents 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 and resource — 'Queries SIDRA tables (IBGE's Automatic Recovery System)' — and immediately distinguishes itself from siblings: 'ibge_sidra is the low-level engine. Prefer a friendlier wrapper when it fits,' naming ibge_censo, ibge_indicadores, ibge_comparar, and ibge_cidades. An agent can tell this tool apart from the rest of the 22-tool family from the description alone.

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-to-use routing with conditions: Census themes → ibge_censo, economic/social time series → ibge_indicadores, rank/compare 2–10 localities → ibge_comparar, one-municipality panel → ibge_cidades, and directs the agent to ibge_sidra_tabelas and ibge_sidra_metadados before querying. It also defines when the estatisticas=true mode is the right choice: 'for largest/smallest/mean/median/distribution/ranking questions'.

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