search_messages

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

Annotations already cover the safety profile (readOnly, non-destructive, open-world), but the description adds substantive behavior: no read-state mutation, provider-specific date semantics (Gmail UTC vs IMAP server calendar days), the pagination invariant that filters/account/mailbox must stay identical across cursor calls (otherwise results are inconsistent), and a prompt-injection warning that content is untrusted. That is meaningful disclosure beyond the annotations, though throttling and result-shape behavior remain unstated.

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 single most important distinction ('Search without marking read'), then dense telegraphic sentences that each carry a distinct constraint. Efficient overall, though the fragment style makes it slightly harder to scan than a structured list would be.

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 10-parameter search tool with 0% schema coverage and no output schema, the description covers the non-obvious parameters and a few real behavioral constraints, but omits matching semantics for the free-text/sender/subject filters and anything about result content or ordering. Adequate but with clear gaps an agent must fill by trial.

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 carry all parameter meaning, and it only partially does. It explains date format/timezone semantics for after/before, the mailbox-as-label-ID rule with the empty-string edge case, and the cursor contract, but leaves text, sender, subject, limit and unread undocumented — the agent must guess whether 'text' is body/substring and how 'sender'/'subject' match.

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 specific verb and resource ('Search...messages') and immediately adds the distinguishing side-effect constraint 'without marking read', which separates it from mark_read and read_message. It is clear what the tool does, though it never names the sibling alternatives explicitly.

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 'without marking read' clause implies when to prefer this over mark_read/read_message, and 'Use next_cursor with identical filters/account/mailbox' gives concrete continuation guidance. However, there is no explicit when-to-use rule or statement of exclusions, so usage is inferred rather than stated.

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