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

Annotations cover the non-destructive, non-open-world profile, and the description adds genuinely useful behavior beyond them: session requirement, the fact OCR is not started, HTTP preference for large payloads, the 201 status, quota/size limits, and dedup of identical files. One tension worth noting is that 'an identical file is reused' hints at repeat-safe behavior while annotations declare idempotentHint=false, though the reuse may apply only to stored content rather than the attachment record, so this is not a clear contradiction.

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 action, then packs constraints into short telegraphic sentences with no filler. The density is high but every clause (OCR, session, HTTP, limits, dedup) carries operational meaning, so it stays efficient rather than padded.

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 mutation tool with no output schema, the description supplies the essentials: authentication/session need, size and quota limits, processing semantics, and the 201 return. Only the response body shape is left unspecified, which is a minor gap given no output schema exists to fill it.

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 all four parameters (cnpj, nome, tipo, arquivo_base64) are already documented in the schema. The description restates the 8 MiB decoded PDF limit but adds no format, encoding, or validation detail beyond what the schema provides, so the baseline of 3 applies.

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 concrete verb and resource (attach a company PDF document) and distinguishes itself from reading-oriented siblings by explicitly saying it does not start OCR and that the automatic reading only happens on an analysis request. An agent can separate this upload tool from get_document / company_documents / analyze_participation without opening any 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?

Gives clear operating context: requires a session, prefer the HTTP path for large files, OCR is not triggered, and analysis is only performed on request. It routes the agent away from expecting immediate processing, but it does not state exclusions relative to siblings such as remove_company_document or list_document_files.

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