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

Annotations already declare readOnlyHint, idempotentHint, and destructiveHint=false, and the trailing 'Read-only' merely restates them. What the description adds beyond annotations is genuinely useful: data provenance (Maple's own customers, not all restaurants), a stated gap (no missed-call rate), and the shape of returned figures (window, cohort size, middle-half spread). It stops short of describing return format structure or any freshness/cadence of the published benchmarks.

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 conditional trigger before the enumeration of returned metrics, and no sentence is filler. The second sentence is dense with comma-separated figure types, which is acceptable but could be tightened into a short list.

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?

A zero-parameter, no-output-schema lookup tool needs its scope, caveats, and returned content explained, and this description does all three, including explicitly flagging the one metric it does not cover. Nothing material is missing for an agent to call 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?

Zero parameters, so there is nothing for the schema to under-document and the baseline applies. The description usefully clarifies that the figures are scoped to a stated window and cohort rather than being timeless constants, which frames how the no-argument result should be read.

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+resource ('what is normal for restaurant phone calls') and enumerates the exact figures returned: calls per day, busiest hours/dayparts/weekdays, after-hours volume, order share, and typical order value. It is clearly distinct from siblings like estimate_missed_call_revenue, which computes a projection rather than reporting published statistics.

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 a concrete trigger: 'Use when a restaurant owner asks what is normal.' It also bounds applicability by noting the data covers Maple's own customers and excludes missed-call rate, which implicitly steers a missed-call question elsewhere, but it never names the alternative tool explicitly.

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