sender_stats

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

Annotations already declare readOnlyHint and openWorldHint, so the bar is lower. The description nonetheless adds material behavior the annotations don't: the result is wrapped in an UNTRUSTED block because it originates from third parties, with an explicit instruction to treat it as data. It does not mention the cost/latency of scanning up to 2000 emails, which would matter for a heavy scan.

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?

Three sentences, each earning its place: what it counts, why you'd use it, and the safety caveat. The core purpose is front-loaded and there is no redundant phrasing or restatement of the tool name.

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 read-only aggregation tool with no output schema, the description covers the counting unit (address and domain), the default time window, and the untrusted-output handling. The only meaningful gap is output shape -- whether results are a ranked list with counts or two separate sections -- which the agent must discover at call time.

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 (max, top, query, account) are already explained in the schema. The description restates the 30-day default for query but adds no syntax guidance, no interaction between 'max' (scan depth) and 'top' (rows shown), and no note that 'account' accepts a label as well as an address. Baseline 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 specific verb and resource -- 'Counts who sends the most mail ... by address and by domain' -- scoped to a Gmail search. This clearly separates it from siblings like search_emails, newsletters, or list_filters, which return content rather than sender aggregates.

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 concrete when-to-use context ('Useful to decide what to filter or label') and states the default search window, so an agent knows this is a triage/analysis tool rather than a retrieval one. It does not explicitly name an alternative or exclusion (e.g., 'use search_emails to read the actual messages'), which keeps it short of a 5.

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