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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive), and the description usefully adds that no caller key is needed, that percentages are a share of people whose status was stated, and that this is census self-report rather than the Health Survey. That is genuine context beyond the structured fields; it stops short of describing filters/response shape in detail.

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?

One dense paragraph that front-loads the subject, scope, and measures before the source caveats. Every sentence carries information, though the embedded example-questions clause makes the middle slightly heavy.

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 no-required-param filtering tool with no output schema, the description conveys the returned dimensions (counts and percentages per region and census year) and the source/semantics caveat. An agent has enough to invoke and interpret it, with only minor gaps around exact response granularity.

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%, and the schema already spells out the year options and region fragment matching, so the baseline is 3. The description adds no syntax, format, or edge-case detail beyond what the schema states.

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?

Names a specific subject (cigarette smoking status), a precise geography (NZ regional council), a population (people aged 15+), the measures (counts and percentages by smoker status), and the years available. An agent can immediately tell this apart from the generic statsnz_observations/codelist siblings, which serve any dataset.

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 example questions it answers ("smoking prevalence by region", "ex-smoker share by regional council, 2018 vs 2023") and explicitly disambiguates the source against the NZ Health Survey. It does not, however, point to an alternative sibling tool for other Stats NZ queries, so routing against statsnz_observations is left implicit.

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