mailbox_list_messages

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

Annotations only assert readOnlyHint, so the description carries the rest and does so well: quarantined messages never appear, blocked content is withheld, cursors are bound to one listing, search results come from an index that may lag ~1 minute, and search pages cap at 50. It even flags third-party content as untrusted data rather than instructions.

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 correctly – purpose, then item shape, then paging, then filter semantics, then caveats. It is a dense single block of roughly 200 words, and the filter-operator enumeration sits mid-paragraph rather than in a scannable list, but nearly every sentence carries non-obvious behavior, so little is wasted.

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 10 optional parameters, no output schema and only a readOnlyHint annotation, the description compensates fully: it documents the returned item fields, paging contract, index staleness, search page limits, quarantine behavior and prompt-injection handling. An agent has everything needed to call and consume this tool correctly.

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%, yet the description still adds real semantics: the actual query operators (de:/from:, para:/to:, assunto:/subject:, tem:anexo, depois:/antes:) that the schema only alludes to as 'operators', and the fact that filters run server-side before the page is cut. It does duplicate the schema's own limit defaults/caps, so it is not fully additive.

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?

Opens with a specific verb+resource+scope: 'List or search one page of messages in a mailbox (inbox, drafts or sent), newest first.' The mailbox/folder/paged framing distinguishes it from siblings like list_emails (outbound) and mailbox_read_message (single message), and the return fields are enumerated immediately.

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 explicit operational rules: pass nextCursor as `cursor` with the same mailbox/folder/filters until hasMore is false, a foreign cursor is refused, and restart without a cursor to pick up newly arrived mail. It also states the negative case – a short or empty page does not prove absence, widen filters before concluding – which is exactly the when-not guidance most tools omit.

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