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

Annotations already carry readOnlyHint, idempotentHint, openWorldHint, and destructiveHint=false, so the bar is lower. The description adds useful context beyond those: "a live GET against the public IBGE SIDRA API," the return shape ("Returns Markdown plus a typed structuredContent payload"), and the statistics-mode side effects ("campos/formato are ignored and registros comes empty; agruparPor... ranks groups by descending sum"). It is consistent with the annotations — no contradiction — and the only absent items (rate limits, pagination details) are minor for a public read-only GET.

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 in the opening sentence, followed by compact list-form years and themes that include English translations. The examples are one-line code-style snippets, the statistics-mode paragraph is dense but necessary, and the routing section is a tight three-item bulleted list. Though longer than average, every sentence earns its place — this is information density, not bloat, for a tool with 9 optional parameters.

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 9-parameter, zero-required tool with 100% schema coverage, an output schema, and four annotation hints, the description covers the essential call patterns, statistics-mode behavior, and sibling routing. One minor gap: the description's 'Available themes' list shows only 9 themes while the schema enum allows 17 (missing migracao, indigenas, quilombolas, saneamento, deficiencia, nupcialidade, fecundidade), which could under-suggest valid options. The note about Markdown plus structuredContent payload closes the return-format loop that the output schema does not fully address.

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 — and the description does add value beyond the schema. It provides concrete usage combinations ("Population 2022: ano='2022', tema='populacao'", "Literacy 2010 by state: ano='2010', tema='alfabetizacao', nivel_territorial='3'"), explains the special 'listar' theme value, and clarifies the interaction semantics of estatisticas with agruparPor/topN/campos/formato. These examples teach call patterns the individual parameter descriptions do not.

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 first sentence states a specific verb and resource: "Queries IBGE Demographic Census data (1970-2022)" with a clear time range. It further distinguishes itself from siblings by self-identifying as a "Simplified tool to access census data without knowing SIDRA table codes," and the 'Use a different tool when' section explicitly names ibge_cidades, ibge_comparar, and ibge_sidra as the alternatives it is not. An agent can immediately tell this from its 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?

The description gives explicit when-to-use guidance: "for largest/smallest/mean/median/distribution/ranking questions over census data ... use estatisticas=true." It then gives explicit when-not-to-use guidance naming alternatives and their conditions: "One municipality's current panel (estimate, HDI, GDP) → ibge_cidades; Comparing/ranking localities → ibge_comparar; An arbitrary SIDRA table → ibge_sidra." Both the trigger conditions and the routing targets are stated, leaving nothing to inference.

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