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

No annotations are provided, so the description carries the full behavioral burden — and it excels: it discloses that the sampling is deterministic and reproducible ('a mesma semente devolve sempre os mesmos casos, então seu resultado é reproduzível por terceiros'), that the sample is randomly drawn, and that the tool is free with no API key ('Gratuito, sem chave'). These are exactly the behavioral traits an agent needs to know and could not infer from the schema or 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 (roughly 350 words) but well-structured and front-loaded — the core purpose comes first, followed by methodology, case contents, verification steps, and reporting rules. Every block earns its place, especially the audit-integrity rules that constrain agent behavior. It is slightly long in the operational verification section (the dfp_cia_aberta_<EXERCICIO>.zip download and filtering details), which borders on over-specification, but it is scannable and purposeful.

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 2-parameter tool with no annotations and no output schema, the description is remarkably complete. It effectively supplies a de facto return contract by enumerating what each case contains (empresa, CD_CVM, exercício, indicador, valor publicado, fórmula, conta CVM), explains the required parameter, gives verification workflow, sets behavioral rules, and even points to the issue tracker for bugs. Nothing an agent needs to invoke it correctly is left to guesswork.

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 0%, so the description must compensate. It thoroughly explains the required parameter 'semente' — what it is (any integer), the security constraint (don't accept suggestions), and its behavioral implication (determinism). The optional 'n' (sample size, default 25) receives no explicit explanation, but it is a simple integer with a sensible default visible in the schema, and the text implicitly references sample sizes ('7 divergências em 100'). Strong compensation for the hard parameter, slightly light on 'n'.

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 leads with a specific verb and resource scope: 'AUDITE ESTA BASE. Sorteia casos e devolve o que você precisa para refazer cada número contra o arquivo ORIGINAL da CVM'. This makes crystal clear it is a sampling-audit tool that cross-checks against original CVM filings, not a fetch/query tool. It is also self-evidently distinct from all 13 siblings (dividendos, fatos_contabeis, fii, etc.), which are data-retrieval tools, so an agent cannot confuse them.

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 strong contextual guidance on the intended scenario ('AUDITE ESTA BASE') and rich methodological rules — choose your own seed, never accept someone else's suggestion, publish the denominator and not just '7 divergências', and show the account when claiming an error. However, it stops short of explicitly naming sibling alternatives or stating when NOT to use this tool versus the regular data tools; the distinction is implied rather than spelled out.

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