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

The annotations already establish readOnly, idempotent, and non-destructive behavior, so the description adds genuine behavioral context: `modo` is ignored when `estatisticas=true`, grouped results are ordered by descending sum with `grupos[0]` as the top spender, `limite` truncates with an `aviso`, and filters use partial search. There is no contradiction with the annotations.

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 dense but every clause earns its place: mode selection, output structures, ordering, filtering, caps, and warnings. The core purpose and the statistics-vs-aggregate decision are front-loaded, and the semicolon-driven structure keeps the long text scannable for an agent.

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?

Given the complexity of 11 parameters and multiple interacting modes, the description is complete: it covers all output forms (`estatisticas`, `grupos[]`, `agregado[]`, `despesas[]`), sorting, truncation, partial filters, and how to resolve `codSenador` via a sibling. The output schema exists, so return-value documentation is not required, but the prose still supplies enough structure for correct invocation.

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?

Although schema coverage is 100%, the description adds substantial cross-parameter meaning: how `estatisticas` interacts with `agruparPor`, how `topN` sizes the top/bottom lists, when `modo` is ignored, and what each output shape contains. This goes far beyond the individual field descriptions in 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?

The description opens with a specific object and scope: CEAPS expenses of senators in a given year. It clearly identifies the resource and what operations are available, distinguishing it from the many senate sibling tools about agendas, legislation, payroll, or procurement.

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 explicit decision rules: use `estatisticas=true` for distribution/ranking/median questions, use aggregate modes for totals, and use `modo='detalhe'` for itemized expenses. It even warns that aggregate modes only sum by group and thus cannot reveal distribution or extremes, and it directs the agent to `senado_listar_senadores` for obtaining `codSenador`.

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