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

Annotations only declare readOnlyHint=true, which the description supports without contradiction. The description adds substantial behavioral disclosure: newest-first ordering, the exact fields returned, the meaning of the starred flag, the fact that total_matches_estimate is Gmail's broader estimate, the no-account error condition, and the rare header-only limitation. This far exceeds what annotations alone provide.

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 longer than average, but each sentence carries distinct operational value: return format, sorting, query syntax, starred-flag semantics, count estimation, and error handling. The most critical routing information (what it returns and how to query) is front-loaded, followed by edge cases. No filler or tautology is present.

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 no output schema present, the description carries the full burden of explaining the return shape and behavior, and it does so thoroughly. It covers result fields, ordering, search syntax, the count estimate, connection-failure handling, and the header-only limitation. For a search tool with Gmail's complex query language, this is complete enough for an agent to select and invoke it 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?

Although schema coverage is 100%, the description meaningfully enriches all three parameters: it explains Gmail search syntax for query with multiple examples, clarifies that omitting account searches all connected mailboxes, and ties max_results to a real example ('last 5 messages ... max_results 5'). It also adds the default/max behavior context in the schema and practical usage scenarios.

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 precise verb and resource: 'Search the user's connected Gmail mailboxes and return matching messages, newest first.' It enumerates the exact return fields (from, to, subject, date, snippet) and clearly distinguishes itself from sibling write tools like archive_gmail_message and create_gmail_draft by focusing on read-only search. It also differentiates from get_gmail_message by indicating this returns a list of messages with snippets rather than a single full message.

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 provides concrete when-to-use guidance with examples: query syntax for sent/received mail, operators like newer_than and has:attachment, and how to phrase natural-language requests. It explicitly explains when to use total_matches_estimate for count questions and gives handling instructions for no-connected-account and header-only-connection cases, including the alternative of answering from the brain. This is rich routing context.

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