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

Annotations already declare readOnlyHint=true and destructiveHint=false, covering safety. The description adds value by disclosing behavioral traits beyond annotations: it mentions that rows carry value, visitors, revenue, and percentage with pagination.total, that limit defaults to 100 (max 1000), and that all filter_* arguments apply with dates defaulting to the last 30 days. This enriches the agent's understanding of the tool's output and behavior without contradicting annotations.

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 information-dense yet efficient, with every sentence contributing meaning. It front-loads the core purpose, then moves to relational context, output characteristics, defaults, and authentication—all in a logical order. No redundant phrases or filler. The structure is ideal for an agent scanning quickly.

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?

Given the tool's complexity (27 parameters, no required fields), the description is remarkably complete. It covers the output shape (rows with value, visitors, revenue, percentage, pagination.total), default behavior (limit 100, last 30 days), filtering semantics (all filter_* apply), and authentication requirements. An output schema exists, so return details are not needed, and the description adds what the schema does not, such as how to relate to sibling tools and the default pagination behavior.

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 the schema already documents all 27 parameters, including their types and descriptions. The description adds minimal extra meaning beyond the schema—it mentions filter_region usage and that all filter_* arguments apply, but these are also present in the schema (e.g., filter_region is described as 'Filter by region code as returned by get_regions'). With full schema coverage, the baseline of 3 is appropriate, and the description does not significantly enhance parameter understanding.

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 clearly states the tool's function: 'Get visitors grouped by region or state (ISO 3166-2 code such as US-CA), ranked by visitors descending, for a date range.' It also explicitly distinguishes itself from siblings by noting it sits between get_countries (coarser) and get_cities (finer), and explains how it can be combined with filter_country and used to scope other tools via filter_region. This provides a precise, resource-specific purpose that an agent can immediately act on.

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 explicit guidance on when to use this tool versus alternatives: 'Sits between get_countries (coarser) and get_cities (finer); combine with filter_country to list the regions of one country. Pass filter_region to other tools to scope them to one region.' It also notes the authentication requirement (requires websiteId or domain with a workspace token) and defaults, making the decision process clear.

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