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

Beyond the annotations it discloses the transactional contract ('É tudo ou nada'), that the response identifies the offending item and reason, that duplicate names are refused with the existing id, and that deactivation removes the record from lists while keeping history. These are real behavioral traits (atomicity, error shape, conflict handling, write target upCampo) not derivable from readOnlyHint=false / destructiveHint=false / idempotentHint=false.

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?

Front-loaded with the highest-value facts (required fields, write target) and organized into short paragraphs by concern (scope, batch modes, conflict handling, confirmation). It is denser than strictly necessary and mixes operational guidance with usage rules, but every sentence carries information.

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 values need not be re-explained; the description instead covers everything an agent must decide before calling (mode selection, batching, confirmation, conflict handling). Nothing material is missing for a 2-parameter write 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 coverage is 100% so the per-parameter descriptions already do the heavy lifting (baseline 3). The description adds cross-field semantics not obvious from the schema alone: the início/fim pair defining the current safra, the id-vs-no-id rule governing include/alter, and id+desativar combination. No syntax or format beyond what the schema states is added.

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 states a specific verb (gravar/cadastrar) and resource (safras), and usefully scopes them as company-wide ('valem para todas as fazendas dela'), which distinguishes them from farm-scoped siblings like cadastrar_talhao or cadastrar_plantio. It never names a sibling explicitly, so the differentiation is inferred from scope rather than stated, keeping it from 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 explicit when-to-use rules for every mode: no id = include, id = alter only the informed fields, id + desativar = deactivate. It also specifies the batch workflow ('mande tudo o que o usuário ditou de uma vez'), the required confirmation step, and how to react to a duplicate-name rejection (ask whether to alter). Little 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.